[
  {
    "id": "mimic-iv",
    "name": "MIMIC-IV",
    "full_name": "Medical Information Mart for Intensive Care IV",
    "description": "The flagship open critical-care database: de-identified EHR data for patients admitted to the ICU or emergency department of Beth Israel Deaconess Medical Center, Boston (2008–2019). Includes demographics, vitals, labs, medications, procedures, diagnoses and survival, organised in modular hospital (hosp) and ICU (icu) tables.",
    "provider": "MIT Laboratory for Computational Physiology / BIDMC",
    "region": "USA",
    "access": "credentialed",
    "access_note": "PhysioNet credentialed account + CITI 'Data or Specimens Only Research' course + signed DUA. Also queryable on Google BigQuery.",
    "url": "https://physionet.org/content/mimiciv/",
    "paper": {
      "label": "Johnson AEW et al. Sci Data 2023;10:1",
      "url": "https://doi.org/10.1038/s41597-022-01899-x"
    },
    "tags": [
      "EHR",
      "ICU",
      "ED",
      "BigQuery"
    ],
    "category": "icu"
  },
  {
    "id": "mimic-iii",
    "name": "MIMIC-III",
    "full_name": "Medical Information Mart for Intensive Care III",
    "description": "The predecessor of MIMIC-IV covering ~46,000 ICU patients at BIDMC (2001–2012), including bedside charting, labs, medications and >2 million free-text clinical notes. Still the most cited ICU dataset and the basis of many benchmarks.",
    "provider": "MIT Laboratory for Computational Physiology / BIDMC",
    "region": "USA",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + CITI training + DUA.",
    "url": "https://physionet.org/content/mimiciii/",
    "paper": {
      "label": "Johnson AEW et al. Sci Data 2016;3:160035",
      "url": "https://doi.org/10.1038/sdata.2016.35"
    },
    "tags": [
      "EHR",
      "ICU",
      "clinical notes"
    ],
    "category": "icu"
  },
  {
    "id": "mimic-iv-demo",
    "name": "MIMIC-IV Clinical Database Demo",
    "description": "A 100-patient subset of MIMIC-IV with identical structure, released openly so anyone can learn the schema and test code before applying for full access.",
    "provider": "MIT Laboratory for Computational Physiology",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download under the Open Database License; no account needed.",
    "url": "https://physionet.org/content/mimic-iv-demo/",
    "paper": null,
    "tags": [
      "EHR",
      "ICU",
      "teaching",
      "demo"
    ],
    "category": "icu"
  },
  {
    "id": "mimic-iv-ed",
    "name": "MIMIC-IV-ED",
    "description": "Emergency department module of MIMIC-IV with ~425,000 ED stays: triage, vital signs, medication reconciliation, dispensed medications and diagnoses. Links to MIMIC-IV hospital and ICU records.",
    "provider": "MIT Laboratory for Computational Physiology / BIDMC",
    "region": "USA",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + CITI training + DUA.",
    "url": "https://physionet.org/content/mimic-iv-ed/",
    "paper": null,
    "tags": [
      "EHR",
      "ED",
      "triage"
    ],
    "category": "icu"
  },
  {
    "id": "mimic-iv-note",
    "name": "MIMIC-IV-Note",
    "description": "De-identified free-text discharge summaries (~330,000) and radiology reports (~2.3 million) linked to MIMIC-IV patients — a major resource for clinical NLP and LLM research.",
    "provider": "MIT Laboratory for Computational Physiology / BIDMC",
    "region": "USA",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + CITI training + DUA. Data must not be sent to third-party online LLM services.",
    "url": "https://physionet.org/content/mimic-iv-note/",
    "paper": null,
    "tags": [
      "clinical notes",
      "NLP",
      "radiology reports"
    ],
    "category": "icu"
  },
  {
    "id": "mimic-iv-echo",
    "name": "MIMIC-IV-ECHO",
    "description": "Echocardiogram DICOM studies (~500,000 videos) from patients in MIMIC-IV, linkable to their EHR data and notes for multimodal cardiac research.",
    "provider": "MIT Laboratory for Computational Physiology / BIDMC",
    "region": "USA",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + CITI training + DUA.",
    "url": "https://physionet.org/content/mimic-iv-echo/",
    "paper": null,
    "tags": [
      "echocardiography",
      "imaging",
      "multimodal"
    ],
    "category": "icu"
  },
  {
    "id": "eicu-crd",
    "name": "eICU-CRD",
    "full_name": "eICU Collaborative Research Database",
    "description": "Multi-centre ICU data for >200,000 admissions across 208 US hospitals (2014–2015) collected through the Philips eICU telehealth programme. Complements single-centre MIMIC for external validation.",
    "provider": "MIT LCP / Philips Healthcare",
    "region": "USA (multi-centre)",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + CITI training + DUA. An open demo subset is also available.",
    "url": "https://physionet.org/content/eicu-crd/",
    "paper": {
      "label": "Pollard TJ et al. Sci Data 2018;5:180178",
      "url": "https://doi.org/10.1038/sdata.2018.178"
    },
    "tags": [
      "EHR",
      "ICU",
      "multi-centre"
    ],
    "category": "icu"
  },
  {
    "id": "amsterdamumcdb",
    "name": "AmsterdamUMCdb",
    "description": "The first freely accessible European ICU database: ~23,000 ICU admissions at Amsterdam UMC (2003–2016) with high-resolution vitals, ventilator settings, labs and drugs (~1 billion data points).",
    "provider": "Amsterdam UMC / Amsterdam Medical Data Science",
    "region": "Netherlands",
    "access": "application",
    "access_note": "Free; requires a short request, completion of human-subjects training and an End User License Agreement.",
    "url": "https://amsterdammedicaldatascience.nl/",
    "paper": {
      "label": "Thoral PJ et al. Crit Care Med 2021;49:e563",
      "url": "https://doi.org/10.1097/CCM.0000000000004916"
    },
    "tags": [
      "EHR",
      "ICU",
      "Europe"
    ],
    "category": "icu"
  },
  {
    "id": "hirid",
    "name": "HiRID",
    "full_name": "High time-Resolution ICU Dataset",
    "description": "Data for ~34,000 ICU admissions at Bern University Hospital with physiological parameters sampled every 2 minutes — notably higher resolution than most ICU datasets. Developed for early warning of circulatory failure.",
    "provider": "Bern University Hospital / ETH Zürich",
    "region": "Switzerland",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + CITI training + DUA.",
    "url": "https://physionet.org/content/hirid/",
    "paper": {
      "label": "Hyland SL et al. Nat Med 2020;26:364",
      "url": "https://doi.org/10.1038/s41591-020-0789-4"
    },
    "tags": [
      "ICU",
      "high-resolution",
      "Europe"
    ],
    "category": "icu"
  },
  {
    "id": "sicdb",
    "name": "SICdb",
    "full_name": "Salzburg Intensive Care Database",
    "description": "ICU data for ~27,000 admissions at University Hospital Salzburg (2013–2021) including per-minute vital signs, therapies, labs and cost-related information.",
    "provider": "Paracelsus Medical University Salzburg",
    "region": "Austria",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + CITI training + DUA.",
    "url": "https://physionet.org/content/sicdb/",
    "paper": {
      "label": "Rodemund N et al. Intensive Care Med 2023;49:987",
      "url": "https://doi.org/10.1007/s00134-023-07046-3"
    },
    "tags": [
      "ICU",
      "Europe"
    ],
    "category": "icu"
  },
  {
    "id": "pic",
    "name": "PIC",
    "full_name": "Paediatric Intensive Care Database",
    "description": "A large paediatric ICU database from the Children's Hospital of Zhejiang University School of Medicine (2010–2018) covering ~13,000 children, structured similarly to MIMIC.",
    "provider": "Zhejiang University",
    "region": "China",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + DUA.",
    "url": "https://physionet.org/content/picdb/",
    "paper": {
      "label": "Zeng X et al. Sci Data 2020;7:14",
      "url": "https://doi.org/10.1038/s41597-020-0355-4"
    },
    "tags": [
      "ICU",
      "paediatrics",
      "Asia"
    ],
    "category": "icu"
  },
  {
    "id": "vitaldb",
    "name": "VitalDB",
    "description": "High-fidelity intraoperative data from 6,388 surgical patients at Seoul National University Hospital: synchronised waveforms (ECG, arterial pressure, EEG/BIS, ventilator), numeric vitals, anaesthetic drugs and perioperative clinical data.",
    "provider": "Seoul National University Hospital",
    "region": "South Korea",
    "access": "open",
    "access_note": "Open download via vitaldb.net, Python API, or PhysioNet (open access).",
    "url": "https://vitaldb.net/dataset/",
    "paper": {
      "label": "Lee HC et al. Sci Data 2022;9:279",
      "url": "https://doi.org/10.1038/s41597-022-01411-5"
    },
    "tags": [
      "anaesthesia",
      "perioperative",
      "waveforms"
    ],
    "category": "icu"
  },
  {
    "id": "inspire",
    "name": "INSPIRE",
    "description": "Perioperative research dataset of ~130,000 surgical cases from Seoul National University Hospital, combining intraoperative vitals with pre- and post-operative EHR data and outcomes.",
    "provider": "Seoul National University Hospital",
    "region": "South Korea",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + DUA.",
    "url": "https://physionet.org/content/inspire/",
    "paper": {
      "label": "Lim L et al. Sci Data 2024;11:655",
      "url": "https://doi.org/10.1038/s41597-024-03517-4"
    },
    "tags": [
      "perioperative",
      "surgery",
      "Asia"
    ],
    "category": "icu"
  },
  {
    "id": "mc-med",
    "name": "MC-MED",
    "full_name": "Multimodal Clinical Monitoring in the Emergency Department",
    "description": "118,385 adult ED visits at Stanford (2020–2022) pairing continuous vitals and waveforms (ECG, PPG, respiration) with structured EHR data and radiology reports.",
    "provider": "Stanford University",
    "region": "USA",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + CITI training + DUA.",
    "url": "https://physionet.org/content/mc-med/",
    "paper": {
      "label": "Kansal A et al. Sci Data 2025",
      "url": "https://doi.org/10.1038/s41597-025-05419-5"
    },
    "tags": [
      "ED",
      "waveforms",
      "multimodal"
    ],
    "category": "icu"
  },
  {
    "id": "mover",
    "name": "MOVER",
    "full_name": "Medical Informatics Operating Room Vitals and Events Repository",
    "description": "Operating-room data from ~58,000 surgical patients at UC Irvine Medical Center, including intraoperative waveforms, vitals, medications and perioperative EHR records.",
    "provider": "University of California, Irvine",
    "region": "USA",
    "access": "application",
    "access_note": "Free download for anyone who signs the data usage agreement (via mover.ics.uci.edu or the UCI repository).",
    "url": "https://archive.ics.uci.edu/dataset/877",
    "paper": {
      "label": "Samad M et al. JAMIA Open 2023;6:ooad084",
      "url": "https://doi.org/10.1093/jamiaopen/ooad084"
    },
    "tags": [
      "perioperative",
      "anaesthesia",
      "waveforms"
    ],
    "category": "icu"
  },
  {
    "id": "physionet-2019-sepsis",
    "name": "PhysioNet/CinC Challenge 2019 (Sepsis)",
    "description": "Hourly ICU data for ~40,000 patients from three hospital systems, released for early prediction of sepsis. A compact, ready-to-model benchmark.",
    "provider": "PhysioNet / Computing in Cardiology",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download; no account needed.",
    "url": "https://physionet.org/content/challenge-2019/",
    "paper": {
      "label": "Reyna MA et al. Crit Care Med 2020;48:210",
      "url": "https://doi.org/10.1097/CCM.0000000000004145"
    },
    "tags": [
      "ICU",
      "sepsis",
      "benchmark"
    ],
    "category": "icu"
  },
  {
    "id": "nwicu",
    "name": "NWICU",
    "full_name": "Northwestern ICU Database",
    "description": "Harmonised ICU data for 25,000+ patients at Northwestern Medicine (2020–2022) in a MIMIC-IV-compatible structure, enabling multi-centre validation.",
    "provider": "Northwestern University",
    "region": "USA",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + CITI training + DUA.",
    "url": "https://physionet.org/content/nwicu-northwestern-icu/",
    "paper": null,
    "tags": [
      "EHR",
      "ICU",
      "MIMIC-compatible"
    ],
    "category": "icu"
  },
  {
    "id": "zigong-icu-infection",
    "name": "Zigong ICU Infection Database",
    "full_name": "Critical care database of patients with infection, Zigong Fourth People's Hospital",
    "description": "ICU records of patients with infection at Zigong Fourth People's Hospital, China (2019–2020), with vitals, labs, drugs and outcomes.",
    "provider": "Zigong Fourth People's Hospital",
    "region": "China",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + DUA.",
    "url": "https://physionet.org/content/icu-infection-zigong-fourth/",
    "paper": null,
    "tags": [
      "ICU",
      "infection",
      "Asia"
    ],
    "category": "icu"
  },
  {
    "id": "mimic-iv-ext-22mcts",
    "name": "MIMIC-IV-Ext-22MCTS",
    "description": "22.6 million clinical time-series events derived from MIMIC-IV discharge summaries for temporal reasoning research.",
    "provider": "PhysioNet",
    "region": "USA",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + DUA.",
    "url": "https://physionet.org/content/mimic-iv-ext-22mcts/",
    "paper": null,
    "tags": [
      "time series",
      "MIMIC",
      "NLP"
    ],
    "category": "icu"
  },
  {
    "id": "inspect",
    "name": "INSPECT",
    "description": "Multimodal pulmonary-embolism dataset of 19,438 patients: CT pulmonary angiograms, radiology impressions and longitudinal structured EHR data with outcome labels.",
    "provider": "Stanford AIMI",
    "region": "USA",
    "access": "registration",
    "access_note": "Free registration and Stanford research-use agreement.",
    "url": "https://stanfordaimi.azurewebsites.net/datasets/521d0d11-7c3c-45ef-8b99-a49cd70cba2c",
    "paper": {
      "label": "Huang SC et al. NeurIPS 2023 Datasets & Benchmarks",
      "url": "https://arxiv.org/abs/2311.10798"
    },
    "tags": [
      "CT",
      "EHR",
      "pulmonary embolism"
    ],
    "category": "icu"
  },
  {
    "id": "uq-vital-signs",
    "name": "University of Queensland Vital Signs Dataset",
    "description": "High-resolution anaesthesia monitoring data (vital signs and waveforms) from 32 surgical cases at the Royal Adelaide Hospital.",
    "provider": "University of Queensland",
    "region": "Australia",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://outbox.eait.uq.edu.au/uqdliu3/uqvitalsignsdataset/index.html",
    "paper": {
      "label": "Liu D et al. Anesth Analg 2012;114:584",
      "url": "https://doi.org/10.1213/ANE.0b013e318241f7c0"
    },
    "tags": [
      "anaesthesia",
      "waveforms"
    ],
    "category": "icu"
  },
  {
    "id": "hdhi-admissions",
    "name": "HDHI Hospital Admissions",
    "description": "14,845 admissions of 12,238 patients to the cardiac unit of Hero DMC Heart Institute, Ludhiana, India (2017–2019) with labs, comorbidities and outcomes.",
    "provider": "Hero DMC Heart Institute",
    "region": "India",
    "access": "registration",
    "access_note": "Free Kaggle account.",
    "url": "https://www.kaggle.com/datasets/ashishsahani/hospital-admissions-data",
    "paper": null,
    "tags": [
      "cardiology",
      "admissions",
      "South Asia"
    ],
    "category": "icu"
  },
  {
    "id": "all-of-us",
    "name": "All of Us Research Hub",
    "description": "NIH programme building a cohort of 1 million+ diverse US participants with EHRs, surveys, physical measurements, wearables and whole-genome sequences, analysed in a cloud Researcher Workbench.",
    "provider": "National Institutes of Health",
    "region": "USA",
    "access": "registration",
    "access_note": "Public Data Browser is open. Registered/Controlled tiers need institutional data use agreement, identity verification and training; cloud credits are provided free initially.",
    "url": "https://www.researchallofus.org/",
    "paper": {
      "label": "All of Us Investigators. N Engl J Med 2019;381:668",
      "url": "https://doi.org/10.1056/NEJMsr1809937"
    },
    "tags": [
      "EHR",
      "genomics",
      "wearables",
      "cohort"
    ],
    "category": "cohorts"
  },
  {
    "id": "n3c",
    "name": "N3C",
    "full_name": "National COVID Cohort Collaborative",
    "description": "A secure enclave harmonising EHR data from >20 million patients (COVID-19 positive and controls) across US health systems into the OMOP model.",
    "provider": "NIH National Center for Advancing Translational Sciences",
    "region": "USA",
    "access": "application",
    "access_note": "Free; institutional data use agreement + data use request per project. Analysis only inside the enclave.",
    "url": "https://ncats.nih.gov/research/research-activities/n3c",
    "paper": {
      "label": "Haendel MA et al. JAMIA 2021;28:427",
      "url": "https://doi.org/10.1093/jamia/ocaa196"
    },
    "tags": [
      "EHR",
      "COVID-19",
      "OMOP"
    ],
    "category": "cohorts"
  },
  {
    "id": "synthea",
    "name": "Synthea",
    "description": "Open-source synthetic patient generator plus downloadable synthetic populations (e.g. SyntheticMass) with complete lifetime records in FHIR, C-CDA and CSV — no privacy restrictions.",
    "provider": "MITRE Corporation",
    "region": "Synthetic (US-based models)",
    "access": "open",
    "access_note": "Direct download; Apache 2.0 software.",
    "url": "https://synthea.mitre.org/downloads",
    "paper": {
      "label": "Walonoski J et al. JAMIA 2018;25:230",
      "url": "https://doi.org/10.1093/jamia/ocx079"
    },
    "tags": [
      "synthetic",
      "FHIR",
      "teaching"
    ],
    "category": "cohorts"
  },
  {
    "id": "cms-synpuf",
    "name": "CMS DE-SynPUF",
    "full_name": "Medicare Claims Synthetic Public Use Files",
    "description": "Synthetic Medicare beneficiary, inpatient, outpatient, carrier and prescription-drug claims (2008–2010) that mirror the structure of real CMS claims for training and software development.",
    "provider": "Centers for Medicare & Medicaid Services",
    "region": "USA (synthetic)",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://www.cms.gov/data-research/statistics-trends-and-reports/medicare-claims-synthetic-public-use-files",
    "paper": null,
    "tags": [
      "claims",
      "synthetic",
      "Medicare"
    ],
    "category": "cohorts"
  },
  {
    "id": "uci-diabetes-130",
    "name": "Diabetes 130-US Hospitals (1999–2008)",
    "description": "~100,000 diabetic inpatient encounters from 130 US hospitals with labs, medications and 30-day readmission outcome — a classic teaching and ML benchmark.",
    "provider": "UCI Machine Learning Repository",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download (CC BY 4.0).",
    "url": "https://archive.ics.uci.edu/dataset/296/diabetes+130-us+hospitals+for+years+1999-2008",
    "paper": {
      "label": "Strack B et al. Biomed Res Int 2014:781670",
      "url": "https://doi.org/10.1155/2014/781670"
    },
    "tags": [
      "EHR",
      "readmission",
      "teaching"
    ],
    "category": "cohorts"
  },
  {
    "id": "seer",
    "name": "SEER",
    "full_name": "Surveillance, Epidemiology, and End Results Program",
    "description": "Population-based cancer registry data covering ~48% of the US population: incidence, stage, treatment and survival since 1975, analysable with the free SEER*Stat software.",
    "provider": "US National Cancer Institute",
    "region": "USA",
    "access": "application",
    "access_note": "Free; request access online and agree to the SEER Research Data Use Agreement (usually approved within days).",
    "url": "https://seer.cancer.gov/data/",
    "paper": null,
    "tags": [
      "cancer",
      "registry",
      "survival"
    ],
    "category": "cohorts"
  },
  {
    "id": "nci-cdas",
    "name": "NCI CDAS (PLCO & NLST)",
    "full_name": "Cancer Data Access System",
    "description": "Individual-level data, images and biospecimen information from the PLCO cancer screening trial (~155,000 participants) and the National Lung Screening Trial (~54,000 participants).",
    "provider": "US National Cancer Institute",
    "region": "USA",
    "access": "application",
    "access_note": "Free; project proposal and data transfer agreement via CDAS.",
    "url": "https://cdas.cancer.gov/",
    "paper": {
      "label": "NLST Research Team. N Engl J Med 2011;365:395",
      "url": "https://doi.org/10.1056/NEJMoa1102873"
    },
    "tags": [
      "cancer screening",
      "trial",
      "imaging"
    ],
    "category": "cohorts"
  },
  {
    "id": "biolincc",
    "name": "NHLBI BioLINCC",
    "full_name": "Biologic Specimen and Data Repository Information Coordinating Center",
    "description": "Individual-level data from >250 NHLBI-funded cohorts and trials such as Framingham, ARIC, MESA, CARDIA, WHI and SPRINT — a gold mine for cardiovascular and lung epidemiology.",
    "provider": "US National Heart, Lung, and Blood Institute",
    "region": "USA",
    "access": "application",
    "access_note": "Free; online request with IRB approval/exemption and a Research Materials Distribution Agreement. Some teaching datasets are open.",
    "url": "https://biolincc.nhlbi.nih.gov/",
    "paper": null,
    "tags": [
      "cohort",
      "cardiovascular",
      "trials"
    ],
    "category": "cohorts"
  },
  {
    "id": "niddk-repository",
    "name": "NIDDK Central Repository",
    "description": "Data and biosamples from NIDDK-funded studies in diabetes, obesity, kidney, liver and digestive diseases (e.g. DPP, TODAY, CRIC).",
    "provider": "US National Institute of Diabetes and Digestive and Kidney Diseases",
    "region": "USA",
    "access": "application",
    "access_note": "Free; registration, research plan and data use agreement.",
    "url": "https://repository.niddk.nih.gov/",
    "paper": null,
    "tags": [
      "diabetes",
      "kidney",
      "cohort"
    ],
    "category": "cohorts"
  },
  {
    "id": "ppmi",
    "name": "PPMI",
    "full_name": "Parkinson's Progression Markers Initiative",
    "description": "Longitudinal clinical, imaging (DaTscan, MRI), genetic and biomarker data from thousands of people with and at risk of Parkinson's disease.",
    "provider": "Michael J. Fox Foundation",
    "region": "International",
    "access": "application",
    "access_note": "Free; online application and Data Use Agreement.",
    "url": "https://www.ppmi-info.org/access-data-specimens/download-data",
    "paper": {
      "label": "Marek K et al. Prog Neurobiol 2011;95:629",
      "url": "https://doi.org/10.1016/j.pneurobio.2011.09.005"
    },
    "tags": [
      "neurology",
      "Parkinson's",
      "cohort"
    ],
    "category": "cohorts"
  },
  {
    "id": "nacc",
    "name": "NACC",
    "full_name": "National Alzheimer's Coordinating Center",
    "description": "Standardised longitudinal clinical and neuropathology data from US Alzheimer's Disease Research Centers (>50,000 participants).",
    "provider": "University of Washington / NIA",
    "region": "USA",
    "access": "application",
    "access_note": "Free; data request with signed Data Use Agreement.",
    "url": "https://naccdata.org/",
    "paper": null,
    "tags": [
      "dementia",
      "neurology",
      "cohort"
    ],
    "category": "cohorts"
  },
  {
    "id": "pro-act",
    "name": "PRO-ACT",
    "full_name": "Pooled Resource Open-Access ALS Clinical Trials",
    "description": "Harmonised records of >11,000 ALS patients pooled from completed clinical trials — the largest ALS clinical trial dataset.",
    "provider": "Healey Center, Massachusetts General Hospital",
    "region": "International",
    "access": "registration",
    "access_note": "Free registration and acceptance of data use terms.",
    "url": "https://ncri1.partners.org/ProACT",
    "paper": {
      "label": "Atassi N et al. Neurology 2014;83:1719",
      "url": "https://doi.org/10.1212/WNL.0000000000000951"
    },
    "tags": [
      "ALS",
      "neurology",
      "trials"
    ],
    "category": "cohorts"
  },
  {
    "id": "uci-heart-disease",
    "name": "UCI Heart Disease (Cleveland)",
    "description": "Small classic dataset of 303 patients from Cleveland (plus Hungary, Switzerland, Long Beach) with 14 clinical attributes for predicting coronary artery disease — widely used in teaching.",
    "provider": "UCI Machine Learning Repository",
    "region": "USA / Europe",
    "access": "open",
    "access_note": "Direct download (CC BY 4.0).",
    "url": "https://archive.ics.uci.edu/dataset/45/heart+disease",
    "paper": null,
    "tags": [
      "cardiology",
      "teaching",
      "tabular"
    ],
    "category": "cohorts"
  },
  {
    "id": "ai-readi",
    "name": "AI-READI",
    "description": "Flagship Bridge2AI dataset on type 2 diabetes: 2,280 participants with retinal imaging, ECG, continuous glucose monitoring, wearables, environmental sensors and surveys (~3.9 TB).",
    "provider": "AI-READI Consortium (NIH Bridge2AI)",
    "region": "USA",
    "access": "registration",
    "access_note": "Free; controlled-access download via FAIRhub after accepting the licence.",
    "url": "https://docs.aireadi.org/",
    "paper": null,
    "tags": [
      "diabetes",
      "multimodal",
      "AI-ready"
    ],
    "category": "cohorts"
  },
  {
    "id": "udn",
    "name": "Undiagnosed Diseases Network (UDN)",
    "description": "Genomic sequences and deep phenotypes of patients with undiagnosed rare conditions, released annually through dbGaP.",
    "provider": "NIH Common Fund / Harvard",
    "region": "USA",
    "access": "application",
    "access_note": "Free; dbGaP authorised-access application.",
    "url": "https://undiagnosed.hms.harvard.edu/research/data-availability/",
    "paper": null,
    "tags": [
      "rare disease",
      "genomics"
    ],
    "category": "cohorts"
  },
  {
    "id": "isaric-covid",
    "name": "ISARIC COVID-19 Data Platform",
    "description": "Standardised clinical data on 700,000+ hospitalised and outpatient COVID-19 patients from 60+ countries.",
    "provider": "ISARIC / Infectious Diseases Data Observatory",
    "region": "Global",
    "access": "application",
    "access_note": "Free; data access request to the IDDO committee.",
    "url": "https://www.iddo.org/covid-19",
    "paper": null,
    "tags": [
      "COVID-19",
      "clinical cohort"
    ],
    "category": "cohorts"
  },
  {
    "id": "ohiot1dm",
    "name": "OhioT1DM",
    "description": "Eight weeks of CGM glucose, insulin, meals, exercise, sleep and wearable data for 12 people with type 1 diabetes.",
    "provider": "Ohio University",
    "region": "USA",
    "access": "application",
    "access_note": "Free; sign and return a data use agreement.",
    "url": "https://webpages.charlotte.edu/rbunescu/data/ohiot1dm/OhioT1DM-dataset.html",
    "paper": {
      "label": "Marling C, Bunescu R. KDH Workshop 2020",
      "url": "https://ceur-ws.org/Vol-2675/paper11.pdf"
    },
    "tags": [
      "diabetes",
      "CGM",
      "wearables"
    ],
    "category": "cohorts"
  },
  {
    "id": "nhanes",
    "name": "NHANES",
    "full_name": "National Health and Nutrition Examination Survey",
    "description": "Continuous US survey combining interviews, physical examinations and laboratory tests (~5,000 people per year) on nutrition, chronic disease, biomarkers and environmental exposures.",
    "provider": "CDC National Center for Health Statistics",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download of public-use files (SAS XPT). Restricted variables via the NCHS Research Data Center.",
    "url": "https://wwwn.cdc.gov/nchs/nhanes/",
    "paper": null,
    "tags": [
      "survey",
      "nutrition",
      "biomarkers"
    ],
    "category": "surveys"
  },
  {
    "id": "brfss",
    "name": "BRFSS",
    "full_name": "Behavioral Risk Factor Surveillance System",
    "description": "The world's largest continuously conducted telephone health survey (>400,000 US adults per year) on risk behaviours, chronic conditions and preventive-service use.",
    "provider": "US Centers for Disease Control and Prevention",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://www.cdc.gov/brfss/annual_data/annual_data.htm",
    "paper": null,
    "tags": [
      "survey",
      "risk factors"
    ],
    "category": "surveys"
  },
  {
    "id": "nhis",
    "name": "NHIS",
    "full_name": "National Health Interview Survey",
    "description": "Annual cross-sectional household survey on the health status, health-care access and insurance of the US civilian population since 1957.",
    "provider": "CDC National Center for Health Statistics",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download of public-use files.",
    "url": "https://www.cdc.gov/nchs/nhis/",
    "paper": null,
    "tags": [
      "survey",
      "access to care"
    ],
    "category": "surveys"
  },
  {
    "id": "meps",
    "name": "MEPS",
    "full_name": "Medical Expenditure Panel Survey",
    "description": "Panel survey of US families, their medical providers and employers detailing health-care use, expenditures, insurance and prescribed medicines.",
    "provider": "Agency for Healthcare Research and Quality",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://meps.ahrq.gov/mepsweb/",
    "paper": null,
    "tags": [
      "survey",
      "health economics"
    ],
    "category": "surveys"
  },
  {
    "id": "yrbss",
    "name": "YRBSS",
    "full_name": "Youth Risk Behavior Surveillance System",
    "description": "Biennial school-based survey of US high-school students on violence, substance use, sexual behaviour, diet, physical activity and mental health.",
    "provider": "US Centers for Disease Control and Prevention",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://www.cdc.gov/yrbs/data/",
    "paper": null,
    "tags": [
      "survey",
      "adolescents"
    ],
    "category": "surveys"
  },
  {
    "id": "nsduh",
    "name": "NSDUH",
    "full_name": "National Survey on Drug Use and Health",
    "description": "Annual nationally representative US survey of tobacco, alcohol and drug use, mental illness and treatment, with public-use files and an online analysis tool.",
    "provider": "SAMHSA",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download via the SAMHSA Data Archive.",
    "url": "https://www.datafiles.samhsa.gov/",
    "paper": null,
    "tags": [
      "survey",
      "substance use",
      "mental health"
    ],
    "category": "surveys"
  },
  {
    "id": "add-health",
    "name": "Add Health",
    "full_name": "National Longitudinal Study of Adolescent to Adult Health",
    "description": "Nationally representative cohort of US adolescents (grades 7–12 in 1994–95) followed into adulthood with social, behavioural, biomarker and genetic data across five waves.",
    "provider": "Carolina Population Center, UNC",
    "region": "USA",
    "access": "registration",
    "access_note": "Public-use files free via ICPSR (login). Restricted-use files require a contract.",
    "url": "https://addhealth.cpc.unc.edu/data/",
    "paper": {
      "label": "Harris KM et al. Int J Epidemiol 2019;48:1415",
      "url": "https://doi.org/10.1093/ije/dyz115"
    },
    "tags": [
      "cohort",
      "adolescents",
      "life course"
    ],
    "category": "surveys"
  },
  {
    "id": "dhs",
    "name": "DHS Program",
    "full_name": "Demographic and Health Surveys",
    "description": "Nationally representative household surveys from 90+ low- and middle-income countries (incl. Pakistan, India's NFHS, Bangladesh, Nigeria) on fertility, maternal and child health, nutrition, HIV and malaria, with GPS cluster data.",
    "provider": "ICF / USAID",
    "region": "Global (LMICs)",
    "access": "application",
    "access_note": "Free; register and submit a short project description — approval typically within 1–2 days.",
    "url": "https://dhsprogram.com/data/",
    "paper": {
      "label": "Corsi DJ et al. Int J Epidemiol 2012;41:1602",
      "url": "https://doi.org/10.1093/ije/dys184"
    },
    "tags": [
      "survey",
      "maternal & child health",
      "LMIC"
    ],
    "category": "surveys"
  },
  {
    "id": "mics",
    "name": "UNICEF MICS",
    "full_name": "Multiple Indicator Cluster Surveys",
    "description": "Household surveys in 100+ countries producing internationally comparable indicators on children and women: nutrition, immunisation, WASH, education and child protection.",
    "provider": "UNICEF",
    "region": "Global (LMICs)",
    "access": "registration",
    "access_note": "Free registration; most datasets downloadable immediately.",
    "url": "https://mics.unicef.org/surveys",
    "paper": null,
    "tags": [
      "survey",
      "child health",
      "LMIC"
    ],
    "category": "surveys"
  },
  {
    "id": "who-ncd-microdata",
    "name": "WHO NCD Microdata Repository (STEPS)",
    "description": "Microdata from WHO STEPwise surveys of non-communicable disease risk factors (tobacco, diet, blood pressure, glucose) in dozens of countries, plus GSHS school surveys.",
    "provider": "World Health Organization",
    "region": "Global",
    "access": "application",
    "access_note": "Free; registration plus a brief data-access request per dataset.",
    "url": "https://extranet.who.int/ncdsmicrodata/index.php/catalog",
    "paper": null,
    "tags": [
      "survey",
      "NCDs",
      "risk factors"
    ],
    "category": "surveys"
  },
  {
    "id": "hrs",
    "name": "HRS",
    "full_name": "Health and Retirement Study",
    "description": "Longitudinal panel of ~20,000 Americans over 50 tracking health, cognition, work, wealth and family since 1992; the template for a global family of ageing studies.",
    "provider": "University of Michigan",
    "region": "USA",
    "access": "registration",
    "access_note": "Free registration for public files; sensitive files by application.",
    "url": "https://hrs.isr.umich.edu/data-products",
    "paper": {
      "label": "Sonnega A et al. Int J Epidemiol 2014;43:576",
      "url": "https://doi.org/10.1093/ije/dyu067"
    },
    "tags": [
      "ageing",
      "cohort",
      "cognition"
    ],
    "category": "surveys"
  },
  {
    "id": "elsa",
    "name": "ELSA",
    "full_name": "English Longitudinal Study of Ageing",
    "description": "Biennial panel of people aged 50+ in England since 2002 covering health, biomarkers, cognition, economics and social participation; harmonised with HRS.",
    "provider": "UCL / UK Data Service",
    "region": "United Kingdom",
    "access": "registration",
    "access_note": "Free registration with the UK Data Service and acceptance of the End User Licence.",
    "url": "https://www.elsa-project.ac.uk/accessing-elsa-data",
    "paper": {
      "label": "Steptoe A et al. Int J Epidemiol 2013;42:1640",
      "url": "https://doi.org/10.1093/ije/dys168"
    },
    "tags": [
      "ageing",
      "cohort"
    ],
    "category": "surveys"
  },
  {
    "id": "share",
    "name": "SHARE",
    "full_name": "Survey of Health, Ageing and Retirement in Europe",
    "description": "Cross-national panel of ~140,000 people aged 50+ in 28 European countries and Israel on health, socio-economic status and social networks.",
    "provider": "SHARE-ERIC",
    "region": "Europe",
    "access": "registration",
    "access_note": "Free for scientific use after registration and acceptance of conditions.",
    "url": "https://share-eric.eu/data/data-access",
    "paper": {
      "label": "Börsch-Supan A et al. Int J Epidemiol 2013;42:992",
      "url": "https://doi.org/10.1093/ije/dyt088"
    },
    "tags": [
      "ageing",
      "cohort",
      "Europe"
    ],
    "category": "surveys"
  },
  {
    "id": "charls",
    "name": "CHARLS",
    "full_name": "China Health and Retirement Longitudinal Study",
    "description": "Nationally representative survey of Chinese residents aged 45+ (~19,000 people) with health, biomarkers, function, income and family data since 2011.",
    "provider": "Peking University",
    "region": "China",
    "access": "registration",
    "access_note": "Free registration and data-use agreement online.",
    "url": "https://charls.pku.edu.cn/en/",
    "paper": {
      "label": "Zhao Y et al. Int J Epidemiol 2014;43:61",
      "url": "https://doi.org/10.1093/ije/dys203"
    },
    "tags": [
      "ageing",
      "cohort",
      "Asia"
    ],
    "category": "surveys"
  },
  {
    "id": "chns",
    "name": "CHNS",
    "full_name": "China Health and Nutrition Survey",
    "description": "Longitudinal household survey (1989 onward) on nutrition, health, economics and the effects of social change across Chinese provinces.",
    "provider": "UNC Carolina Population Center / China CDC",
    "region": "China",
    "access": "registration",
    "access_note": "Free registration to download.",
    "url": "https://www.cpc.unc.edu/projects/china",
    "paper": {
      "label": "Popkin BM et al. Int J Epidemiol 2010;39:1435",
      "url": "https://doi.org/10.1093/ije/dyp322"
    },
    "tags": [
      "nutrition",
      "cohort",
      "Asia"
    ],
    "category": "surveys"
  },
  {
    "id": "knhanes",
    "name": "KNHANES",
    "full_name": "Korea National Health and Nutrition Examination Survey",
    "description": "South Korea's national health-examination survey (since 1998) with interviews, physical exams, labs and nutrition data, modelled on NHANES.",
    "provider": "Korea Disease Control and Prevention Agency",
    "region": "South Korea",
    "access": "registration",
    "access_note": "Free registration and agreement to the data-use pledge.",
    "url": "https://knhanes.kdca.go.kr/knhanes/main.do",
    "paper": {
      "label": "Kweon S et al. Int J Epidemiol 2014;43:69",
      "url": "https://doi.org/10.1093/ije/dyt228"
    },
    "tags": [
      "survey",
      "nutrition",
      "Asia"
    ],
    "category": "surveys"
  },
  {
    "id": "g2aging",
    "name": "Gateway to Global Aging Data",
    "description": "Harmonised data files and documentation for the HRS family of ageing studies (HRS, ELSA, SHARE, CHARLS, LASI, MHAS, KLoSA and more) for cross-country comparison.",
    "provider": "University of Southern California",
    "region": "Global",
    "access": "registration",
    "access_note": "Free registration; underlying data obtained from each study.",
    "url": "https://g2aging.org/",
    "paper": null,
    "tags": [
      "ageing",
      "harmonised",
      "cross-national"
    ],
    "category": "surveys"
  },
  {
    "id": "ipums-health",
    "name": "IPUMS Health Surveys",
    "description": "Harmonised, documented versions of NHIS (1963–present) and MEPS with a custom data-extract builder that makes multi-year analysis simple.",
    "provider": "University of Minnesota",
    "region": "USA",
    "access": "registration",
    "access_note": "Free registration.",
    "url": "https://healthsurveys.ipums.org/",
    "paper": null,
    "tags": [
      "survey",
      "harmonised"
    ],
    "category": "surveys"
  },
  {
    "id": "elsi-brazil",
    "name": "ELSI-Brazil",
    "full_name": "Brazilian Longitudinal Study of Ageing",
    "description": "Nationally representative cohort of Brazilians aged 50+ (9,412 at baseline) covering health, function, care and economics; part of the HRS family.",
    "provider": "Fiocruz / UFMG",
    "region": "Brazil",
    "access": "registration",
    "access_note": "Free registration and acceptance of terms.",
    "url": "https://elsi.cpqrr.fiocruz.br/en/home-english/en-data-access/",
    "paper": {
      "label": "Lima-Costa MF et al. Am J Epidemiol 2018;187:1345",
      "url": "https://doi.org/10.1093/aje/kwx387"
    },
    "tags": [
      "ageing",
      "cohort",
      "Latin America"
    ],
    "category": "surveys"
  },
  {
    "id": "nanda",
    "name": "NaNDA",
    "full_name": "National Neighborhood Data Archive",
    "description": "US neighbourhood-level measures (food, healthcare, transport, pollution, amenities) by census tract and ZIP code for social-determinants research.",
    "provider": "University of Michigan / ICPSR",
    "region": "USA",
    "access": "registration",
    "access_note": "Free ICPSR login.",
    "url": "https://www.icpsr.umich.edu/web/ICPSR/series/1920",
    "paper": null,
    "tags": [
      "social determinants",
      "geospatial"
    ],
    "category": "surveys"
  },
  {
    "id": "who-gho",
    "name": "WHO Global Health Observatory",
    "description": "WHO's portal of 1,000+ health indicators for 194 member states — mortality, disease burden, health systems, SDG targets — with an OData API.",
    "provider": "World Health Organization",
    "region": "Global",
    "access": "open",
    "access_note": "Open web, CSV and API access.",
    "url": "https://www.who.int/data/gho",
    "paper": null,
    "tags": [
      "indicators",
      "API"
    ],
    "category": "global"
  },
  {
    "id": "gbd",
    "name": "GBD Results (IHME)",
    "full_name": "Global Burden of Disease Study",
    "description": "Estimates of deaths, DALYs, prevalence and risk-factor burden for 370+ diseases and injuries in 204 countries from 1990 onward.",
    "provider": "Institute for Health Metrics and Evaluation",
    "region": "Global",
    "access": "registration",
    "access_note": "Free for non-commercial use; login required for bulk downloads.",
    "url": "https://vizhub.healthdata.org/gbd-results/",
    "paper": {
      "label": "GBD 2021 Diseases and Injuries Collaborators. Lancet 2024;403:2133",
      "url": "https://doi.org/10.1016/S0140-6736(24)00757-8"
    },
    "tags": [
      "burden of disease",
      "DALYs"
    ],
    "category": "global"
  },
  {
    "id": "worldbank-hnp",
    "name": "World Bank Health, Nutrition & Population Statistics",
    "description": "Country-level time series on health financing, service coverage, nutrition, population dynamics and outcomes, downloadable via DataBank or API.",
    "provider": "World Bank",
    "region": "Global",
    "access": "open",
    "access_note": "Open (CC BY 4.0).",
    "url": "https://databank.worldbank.org/source/health-nutrition-and-population-statistics",
    "paper": null,
    "tags": [
      "indicators",
      "health financing"
    ],
    "category": "global"
  },
  {
    "id": "owid",
    "name": "Our World in Data",
    "description": "Curated, well-documented global datasets and charts on health topics (causes of death, vaccination, COVID-19, child mortality, life expectancy) with GitHub-hosted data.",
    "provider": "Global Change Data Lab / University of Oxford",
    "region": "Global",
    "access": "open",
    "access_note": "Open (CC BY 4.0).",
    "url": "https://ourworldindata.org/",
    "paper": {
      "label": "Mathieu E et al. Nat Hum Behav 2021;5:947",
      "url": "https://doi.org/10.1038/s41562-021-01122-8"
    },
    "tags": [
      "indicators",
      "visualisation"
    ],
    "category": "global"
  },
  {
    "id": "who-mortality",
    "name": "WHO Mortality Database",
    "description": "Deaths by country, year, sex, age and ICD cause as reported by national civil-registration systems since 1950.",
    "provider": "World Health Organization",
    "region": "Global",
    "access": "open",
    "access_note": "Open web tool and raw-file download.",
    "url": "https://platform.who.int/mortality",
    "paper": null,
    "tags": [
      "mortality",
      "cause of death"
    ],
    "category": "global"
  },
  {
    "id": "hmd",
    "name": "Human Mortality Database",
    "description": "High-quality, harmonised life tables, death counts and population exposures for ~40 countries, the reference source for demographic mortality research.",
    "provider": "UC Berkeley / Max Planck Institute for Demographic Research",
    "region": "Global (mainly high-income)",
    "access": "registration",
    "access_note": "Free registration.",
    "url": "https://www.mortality.org/",
    "paper": null,
    "tags": [
      "mortality",
      "life tables",
      "demography"
    ],
    "category": "global"
  },
  {
    "id": "gco",
    "name": "Global Cancer Observatory (GLOBOCAN)",
    "description": "IARC's estimates of cancer incidence, mortality and prevalence for 185 countries, plus Cancer Over Time, Cancer Tomorrow and Cancer Survival modules.",
    "provider": "International Agency for Research on Cancer",
    "region": "Global",
    "access": "open",
    "access_note": "Open web tool and downloads.",
    "url": "https://gco.iarc.who.int/",
    "paper": {
      "label": "Bray F et al. CA Cancer J Clin 2024;74:229",
      "url": "https://doi.org/10.3322/caac.21834"
    },
    "tags": [
      "cancer",
      "incidence"
    ],
    "category": "global"
  },
  {
    "id": "cdc-wonder",
    "name": "CDC WONDER",
    "description": "Online query system for US public-health data: births, deaths (multiple cause), cancer statistics, vaccine adverse events, environmental and population data.",
    "provider": "US Centers for Disease Control and Prevention",
    "region": "USA",
    "access": "open",
    "access_note": "Open web queries; agree to data-use restrictions on each query.",
    "url": "https://wonder.cdc.gov/",
    "paper": null,
    "tags": [
      "mortality",
      "natality",
      "query tool"
    ],
    "category": "global"
  },
  {
    "id": "nchs-vital",
    "name": "NCHS Vital Statistics Online",
    "description": "Public-use micro-data files for every US birth, death and fetal death record, released annually by NCHS.",
    "provider": "CDC National Center for Health Statistics",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://www.cdc.gov/nchs/data_access/vitalstatsonline.htm",
    "paper": null,
    "tags": [
      "births",
      "deaths",
      "microdata"
    ],
    "category": "global"
  },
  {
    "id": "cdc-places",
    "name": "CDC PLACES",
    "description": "Model-based small-area estimates of 40 chronic-disease measures for US counties, cities, census tracts and ZIP codes.",
    "provider": "US Centers for Disease Control and Prevention",
    "region": "USA",
    "access": "open",
    "access_note": "Open (Socrata API).",
    "url": "https://www.cdc.gov/places/",
    "paper": null,
    "tags": [
      "small-area",
      "chronic disease",
      "geospatial"
    ],
    "category": "global"
  },
  {
    "id": "unaids",
    "name": "UNAIDS AIDSinfo",
    "description": "Country-level HIV estimates — prevalence, incidence, AIDS deaths, treatment cascade and key populations — from UNAIDS modelling.",
    "provider": "UNAIDS",
    "region": "Global",
    "access": "open",
    "access_note": "Open web tool and downloads.",
    "url": "https://aidsinfo.unaids.org/",
    "paper": null,
    "tags": [
      "HIV",
      "indicators"
    ],
    "category": "global"
  },
  {
    "id": "eurostat-health",
    "name": "Eurostat Health Statistics",
    "description": "EU-wide statistics on causes of death, health-care resources and expenditure, health status (EHIS) and accidents at work.",
    "provider": "Eurostat",
    "region": "Europe",
    "access": "open",
    "access_note": "Open.",
    "url": "https://ec.europa.eu/eurostat/web/health/database",
    "paper": null,
    "tags": [
      "indicators",
      "Europe"
    ],
    "category": "global"
  },
  {
    "id": "healthdata-gov",
    "name": "HealthData.gov",
    "description": "US federal catalogue of thousands of health datasets from HHS agencies (CDC, CMS, FDA, NIH) with API access.",
    "provider": "US Department of Health & Human Services",
    "region": "USA",
    "access": "open",
    "access_note": "Open.",
    "url": "https://healthdata.gov/",
    "paper": null,
    "tags": [
      "catalogue",
      "API"
    ],
    "category": "global"
  },
  {
    "id": "cms-data",
    "name": "CMS Data (data.cms.gov)",
    "description": "Public Medicare and Medicaid datasets: provider utilisation and payments, Part D prescribers, hospital quality and cost reports.",
    "provider": "Centers for Medicare & Medicaid Services",
    "region": "USA",
    "access": "open",
    "access_note": "Open (API and CSV).",
    "url": "https://data.cms.gov/",
    "paper": null,
    "tags": [
      "claims",
      "health services"
    ],
    "category": "global"
  },
  {
    "id": "datasus",
    "name": "DATASUS (TabNet)",
    "description": "Brazilian Ministry of Health information system: hospital admissions (SIH), mortality (SIM), births (SINASC), notifiable diseases (SINAN) and more.",
    "provider": "Brazilian Ministry of Health",
    "region": "Brazil",
    "access": "open",
    "access_note": "Open web tabulation and file downloads.",
    "url": "https://datasus.saude.gov.br/informacoes-de-saude-tabnet/",
    "paper": null,
    "tags": [
      "health services",
      "mortality",
      "Latin America"
    ],
    "category": "global"
  },
  {
    "id": "worldpop",
    "name": "WorldPop",
    "description": "High-resolution gridded population, demographic, births and urbanisation estimates for every country — essential denominators for health mapping.",
    "provider": "University of Southampton",
    "region": "Global",
    "access": "open",
    "access_note": "Open download (CC BY 4.0).",
    "url": "https://www.worldpop.org/datacatalog/",
    "paper": null,
    "tags": [
      "population",
      "geospatial"
    ],
    "category": "global"
  },
  {
    "id": "mimic-cxr",
    "name": "MIMIC-CXR",
    "description": "377,110 chest radiographs (DICOM) with 227,835 free-text radiology reports from BIDMC, linkable to MIMIC-IV. A JPG version with NLP-derived labels (MIMIC-CXR-JPG) is also available.",
    "provider": "MIT Laboratory for Computational Physiology",
    "region": "USA",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + CITI training + DUA.",
    "url": "https://physionet.org/content/mimic-cxr/",
    "paper": {
      "label": "Johnson AEW et al. Sci Data 2019;6:317",
      "url": "https://doi.org/10.1038/s41597-019-0322-0"
    },
    "tags": [
      "chest X-ray",
      "reports",
      "multimodal"
    ],
    "category": "imaging"
  },
  {
    "id": "chexpert",
    "name": "CheXpert",
    "description": "224,316 chest radiographs of 65,240 Stanford patients labelled for 14 observations with uncertainty labels; the basis of a well-known benchmark.",
    "provider": "Stanford ML Group / Stanford AIMI",
    "region": "USA",
    "access": "registration",
    "access_note": "Free registration and acceptance of the Stanford research-use agreement.",
    "url": "https://stanfordmlgroup.github.io/competitions/chexpert/",
    "paper": {
      "label": "Irvin J et al. AAAI 2019;33:590",
      "url": "https://doi.org/10.1609/aaai.v33i01.3301590"
    },
    "tags": [
      "chest X-ray",
      "labels"
    ],
    "category": "imaging"
  },
  {
    "id": "nih-chestxray14",
    "name": "NIH ChestX-ray14",
    "description": "112,120 frontal chest X-rays from 30,805 patients at the NIH Clinical Center with 14 text-mined disease labels and some bounding boxes.",
    "provider": "NIH Clinical Center",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download (Box), no account needed.",
    "url": "https://nihcc.app.box.com/v/ChestXray-NIHCC",
    "paper": {
      "label": "Wang X et al. CVPR 2017",
      "url": "https://doi.org/10.1109/CVPR.2017.369"
    },
    "tags": [
      "chest X-ray",
      "labels"
    ],
    "category": "imaging"
  },
  {
    "id": "padchest",
    "name": "PadChest",
    "description": "160,000+ chest X-rays from Hospital San Juan (Spain) with Spanish reports and 174 radiographic findings, 27% manually annotated by radiologists.",
    "provider": "BIMCV, Valencia",
    "region": "Spain",
    "access": "application",
    "access_note": "Free; request form and research-use agreement.",
    "url": "https://bimcv.cipf.es/bimcv-projects/padchest/",
    "paper": {
      "label": "Bustos A et al. Med Image Anal 2020;66:101797",
      "url": "https://doi.org/10.1016/j.media.2020.101797"
    },
    "tags": [
      "chest X-ray",
      "reports",
      "Europe"
    ],
    "category": "imaging"
  },
  {
    "id": "vindr-cxr",
    "name": "VinDr-CXR",
    "description": "18,000 chest X-rays from two Vietnamese hospitals with radiologist bounding-box annotations for 22 local findings and 6 global diagnoses.",
    "provider": "Vingroup Big Data Institute",
    "region": "Vietnam",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + DUA.",
    "url": "https://physionet.org/content/vindr-cxr/",
    "paper": {
      "label": "Nguyen HQ et al. Sci Data 2022;9:429",
      "url": "https://doi.org/10.1038/s41597-022-01498-w"
    },
    "tags": [
      "chest X-ray",
      "bounding boxes",
      "Asia"
    ],
    "category": "imaging"
  },
  {
    "id": "vindr-mammo",
    "name": "VinDr-Mammo",
    "description": "5,000 four-view full-field digital mammography exams with BI-RADS, density and finding-level annotations from Vietnamese hospitals.",
    "provider": "Vingroup Big Data Institute",
    "region": "Vietnam",
    "access": "registration",
    "access_note": "Free PhysioNet account + signed data use agreement (no credentialing needed).",
    "url": "https://physionet.org/content/vindr-mammo/",
    "paper": {
      "label": "Nguyen HT et al. Sci Data 2023;10:277",
      "url": "https://doi.org/10.1038/s41597-023-02100-7"
    },
    "tags": [
      "mammography",
      "Asia"
    ],
    "category": "imaging"
  },
  {
    "id": "rsna-pneumonia",
    "name": "RSNA Pneumonia Detection Challenge",
    "description": "~30,000 chest X-rays (from NIH ChestX-ray14) re-annotated by radiologists with bounding boxes for lung opacity suggestive of pneumonia.",
    "provider": "Radiological Society of North America",
    "region": "USA",
    "access": "registration",
    "access_note": "Free Kaggle account and acceptance of competition rules.",
    "url": "https://www.kaggle.com/c/rsna-pneumonia-detection-challenge",
    "paper": {
      "label": "Shih G et al. Radiol Artif Intell 2019;1:e180041",
      "url": "https://doi.org/10.1148/ryai.2019180041"
    },
    "tags": [
      "chest X-ray",
      "pneumonia",
      "challenge"
    ],
    "category": "imaging"
  },
  {
    "id": "tcia",
    "name": "The Cancer Imaging Archive (TCIA)",
    "description": "Large archive of de-identified cancer imaging collections (CT, MRI, PET, pathology) — e.g. LIDC-IDRI, CBIS-DDSM, NLST — often linked to TCGA genomics and clinical data.",
    "provider": "US National Cancer Institute / University of Arkansas",
    "region": "International",
    "access": "open",
    "access_note": "Most collections open (CC BY); some limited-access collections require a signed agreement.",
    "url": "https://www.cancerimagingarchive.net/",
    "paper": {
      "label": "Clark K et al. J Digit Imaging 2013;26:1045",
      "url": "https://doi.org/10.1007/s10278-013-9622-7"
    },
    "tags": [
      "oncology",
      "CT",
      "MRI",
      "hub"
    ],
    "category": "imaging"
  },
  {
    "id": "lidc-idri",
    "name": "LIDC-IDRI",
    "description": "1,018 thoracic CT scans with lung-nodule annotations by four radiologists — the standard dataset for lung-nodule detection (LUNA16 derives from it).",
    "provider": "NCI / FNIH via TCIA",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from TCIA.",
    "url": "https://www.cancerimagingarchive.net/collection/lidc-idri/",
    "paper": {
      "label": "Armato SG et al. Med Phys 2011;38:915",
      "url": "https://doi.org/10.1118/1.3528204"
    },
    "tags": [
      "CT",
      "lung nodules"
    ],
    "category": "imaging"
  },
  {
    "id": "cbis-ddsm",
    "name": "CBIS-DDSM",
    "description": "Curated, DICOM-converted subset of the Digital Database for Screening Mammography with updated ROI segmentations and pathology-verified labels.",
    "provider": "TCIA",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from TCIA.",
    "url": "https://www.cancerimagingarchive.net/collection/cbis-ddsm/",
    "paper": {
      "label": "Lee RS et al. Sci Data 2017;4:170177",
      "url": "https://doi.org/10.1038/sdata.2017.177"
    },
    "tags": [
      "mammography",
      "segmentation"
    ],
    "category": "imaging"
  },
  {
    "id": "embed",
    "name": "EMBED",
    "full_name": "Emory Breast Imaging Dataset",
    "description": "3.4 million screening and diagnostic mammograms from 116,000 racially diverse women with BI-RADS, pathology and demographic data.",
    "provider": "Emory University",
    "region": "USA",
    "access": "application",
    "access_note": "Free for research; data-use agreement and AWS access request.",
    "url": "https://registry.opendata.aws/emory-breast-imaging-dataset-embed/",
    "paper": {
      "label": "Jeong JJ et al. Radiol Artif Intell 2023;5:e220047",
      "url": "https://doi.org/10.1148/ryai.220047"
    },
    "tags": [
      "mammography",
      "diversity"
    ],
    "category": "imaging"
  },
  {
    "id": "msd",
    "name": "Medical Segmentation Decathlon",
    "description": "Ten 3D segmentation tasks (brain tumour, heart, liver, hippocampus, prostate, lung, pancreas, hepatic vessels, spleen, colon) for benchmarking generalisable algorithms.",
    "provider": "MSD consortium",
    "region": "International",
    "access": "open",
    "access_note": "Direct download (CC BY-SA 4.0).",
    "url": "http://medicaldecathlon.com/",
    "paper": {
      "label": "Antonelli M et al. Nat Commun 2022;13:4128",
      "url": "https://doi.org/10.1038/s41467-022-30695-9"
    },
    "tags": [
      "segmentation",
      "CT",
      "MRI",
      "benchmark"
    ],
    "category": "imaging"
  },
  {
    "id": "totalsegmentator",
    "name": "TotalSegmentator dataset",
    "description": "1,228 CT scans with segmentations of 117 anatomical structures (organs, bones, muscles, vessels), released alongside the popular TotalSegmentator tool.",
    "provider": "University Hospital Basel",
    "region": "Switzerland",
    "access": "open",
    "access_note": "Direct download from Zenodo (CC BY 4.0).",
    "url": "https://zenodo.org/records/10047292",
    "paper": {
      "label": "Wasserthal J et al. Radiol Artif Intell 2023;5:e230024",
      "url": "https://doi.org/10.1148/ryai.230024"
    },
    "tags": [
      "CT",
      "segmentation",
      "anatomy"
    ],
    "category": "imaging"
  },
  {
    "id": "kits",
    "name": "KiTS",
    "full_name": "Kidney Tumor Segmentation Challenge",
    "description": "Contrast-enhanced abdominal CT scans with expert segmentations of kidneys, tumours and cysts (KiTS19/21/23).",
    "provider": "University of Minnesota",
    "region": "USA",
    "access": "open",
    "access_note": "Download via GitHub script (CC BY-NC-SA).",
    "url": "https://kits-challenge.org/",
    "paper": {
      "label": "Heller N et al. Med Image Anal 2021;67:101821",
      "url": "https://doi.org/10.1016/j.media.2020.101821"
    },
    "tags": [
      "CT",
      "kidney",
      "segmentation"
    ],
    "category": "imaging"
  },
  {
    "id": "lits",
    "name": "LiTS",
    "full_name": "Liver Tumor Segmentation Benchmark",
    "description": "131 training and 70 test contrast-enhanced abdominal CT scans with liver and tumour segmentations from seven institutions.",
    "provider": "LiTS consortium",
    "region": "International",
    "access": "registration",
    "access_note": "Free registration on CodaLab.",
    "url": "https://competitions.codalab.org/competitions/17094",
    "paper": {
      "label": "Bilic P et al. Med Image Anal 2023;84:102680",
      "url": "https://doi.org/10.1016/j.media.2022.102680"
    },
    "tags": [
      "CT",
      "liver",
      "segmentation"
    ],
    "category": "imaging"
  },
  {
    "id": "luna16",
    "name": "LUNA16",
    "description": "888 CT scans from LIDC-IDRI with standardised nodule annotations and an evaluation framework for nodule detection.",
    "provider": "Grand Challenge",
    "region": "International",
    "access": "registration",
    "access_note": "Free Grand Challenge account; data also on Zenodo.",
    "url": "https://luna16.grand-challenge.org/",
    "paper": {
      "label": "Setio AAA et al. Med Image Anal 2017;42:1",
      "url": "https://doi.org/10.1016/j.media.2017.06.015"
    },
    "tags": [
      "CT",
      "lung nodules",
      "challenge"
    ],
    "category": "imaging"
  },
  {
    "id": "fastmri",
    "name": "fastMRI",
    "description": "Raw k-space and DICOM data for knee, brain, prostate and breast MRI from NYU Langone, built for accelerated MRI reconstruction research.",
    "provider": "NYU Langone Health / Meta AI",
    "region": "USA",
    "access": "registration",
    "access_note": "Free; submit a short form and accept the data sharing agreement to receive download links.",
    "url": "https://fastmri.med.nyu.edu/",
    "paper": {
      "label": "Knoll F et al. Radiol Artif Intell 2020;2:e190007",
      "url": "https://doi.org/10.1148/ryai.2020190007"
    },
    "tags": [
      "MRI",
      "reconstruction",
      "raw data"
    ],
    "category": "imaging"
  },
  {
    "id": "echonet-dynamic",
    "name": "EchoNet-Dynamic",
    "description": "10,030 apical four-chamber echocardiography videos with expert tracings, ejection fraction and volumes from Stanford Medicine.",
    "provider": "Stanford AIMI",
    "region": "USA",
    "access": "registration",
    "access_note": "Free registration and research-use agreement (Stanford AIMI).",
    "url": "https://echonet.github.io/dynamic/",
    "paper": {
      "label": "Ouyang D et al. Nature 2020;580:252",
      "url": "https://doi.org/10.1038/s41586-020-2145-8"
    },
    "tags": [
      "echocardiography",
      "video"
    ],
    "category": "imaging"
  },
  {
    "id": "camus",
    "name": "CAMUS",
    "description": "2D echocardiographic sequences from 500 patients with expert segmentations of left ventricle, myocardium and left atrium.",
    "provider": "CREATIS, University of Lyon",
    "region": "France",
    "access": "registration",
    "access_note": "Free registration.",
    "url": "https://www.creatis.insa-lyon.fr/Challenge/camus/",
    "paper": {
      "label": "Leclerc S et al. IEEE Trans Med Imaging 2019;38:2198",
      "url": "https://doi.org/10.1109/TMI.2019.2900516"
    },
    "tags": [
      "echocardiography",
      "segmentation"
    ],
    "category": "imaging"
  },
  {
    "id": "acdc",
    "name": "ACDC",
    "full_name": "Automated Cardiac Diagnosis Challenge",
    "description": "Cine cardiac MRI from 150 patients across five diagnostic groups with ventricular and myocardial segmentations.",
    "provider": "CREATIS, University of Lyon",
    "region": "France",
    "access": "registration",
    "access_note": "Free registration.",
    "url": "https://www.creatis.insa-lyon.fr/Challenge/acdc/",
    "paper": {
      "label": "Bernard O et al. IEEE Trans Med Imaging 2018;37:2514",
      "url": "https://doi.org/10.1109/TMI.2018.2837502"
    },
    "tags": [
      "cardiac MRI",
      "segmentation"
    ],
    "category": "imaging"
  },
  {
    "id": "busi",
    "name": "BUSI",
    "full_name": "Breast Ultrasound Images Dataset",
    "description": "780 breast ultrasound images from 600 women in Egypt, labelled normal, benign or malignant with ground-truth masks.",
    "provider": "Cairo University",
    "region": "Egypt",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://scholar.cu.edu.eg/?q=afahmy/pages/dataset",
    "paper": {
      "label": "Al-Dhabyani W et al. Data Brief 2020;28:104863",
      "url": "https://doi.org/10.1016/j.dib.2019.104863"
    },
    "tags": [
      "ultrasound",
      "breast"
    ],
    "category": "imaging"
  },
  {
    "id": "kermany-oct-cxr",
    "name": "Kermany OCT & Paediatric Chest X-ray",
    "description": "108,000+ labelled retinal OCT images (CNV, DME, drusen, normal) and 5,800 paediatric chest X-rays (pneumonia vs normal) from the landmark Cell transfer-learning paper.",
    "provider": "UC San Diego / Guangzhou Women and Children's Medical Center",
    "region": "USA / China",
    "access": "open",
    "access_note": "Direct download from Mendeley Data (CC BY 4.0).",
    "url": "https://data.mendeley.com/datasets/rscbjbr9sj/2",
    "paper": {
      "label": "Kermany DS et al. Cell 2018;172:1122",
      "url": "https://doi.org/10.1016/j.cell.2018.02.010"
    },
    "tags": [
      "OCT",
      "chest X-ray",
      "paediatrics"
    ],
    "category": "imaging"
  },
  {
    "id": "mura",
    "name": "MURA",
    "description": "40,561 musculoskeletal radiographs of the upper extremity from 12,173 patients labelled normal/abnormal by radiologists.",
    "provider": "Stanford ML Group",
    "region": "USA",
    "access": "registration",
    "access_note": "Free registration and research-use agreement.",
    "url": "https://stanfordmlgroup.github.io/competitions/mura/",
    "paper": {
      "label": "Rajpurkar P et al. arXiv 2017",
      "url": "https://arxiv.org/abs/1712.06957"
    },
    "tags": [
      "musculoskeletal",
      "X-ray"
    ],
    "category": "imaging"
  },
  {
    "id": "stanford-aimi",
    "name": "Stanford AIMI Shared Datasets",
    "description": "Portal hosting many Stanford imaging datasets free for research: CheXpert, EchoNet, DDI, CT colonography, knee MRI, BrainMetShare and more.",
    "provider": "Stanford Center for AI in Medicine & Imaging",
    "region": "USA",
    "access": "registration",
    "access_note": "Free registration and Stanford research-use agreement per dataset.",
    "url": "https://aimi.stanford.edu/shared-datasets",
    "paper": null,
    "tags": [
      "hub",
      "radiology"
    ],
    "category": "imaging"
  },
  {
    "id": "open-i",
    "name": "Open-i (incl. IU X-ray)",
    "description": "NLM search engine over 3.7 million images from PubMed Central articles plus the Indiana University chest X-ray collection (7,470 images with reports).",
    "provider": "US National Library of Medicine",
    "region": "USA",
    "access": "open",
    "access_note": "Open web and API; IU X-ray downloadable.",
    "url": "https://openi.nlm.nih.gov/",
    "paper": {
      "label": "Demner-Fushman D et al. JAMIA 2016;23:304",
      "url": "https://doi.org/10.1093/jamia/ocv080"
    },
    "tags": [
      "chest X-ray",
      "reports",
      "literature images"
    ],
    "category": "imaging"
  },
  {
    "id": "medmnist",
    "name": "MedMNIST v2",
    "description": "18 standardised, lightweight 2D and 3D biomedical image datasets (pathology, chest X-ray, dermatology, retina, CT organs…) in MNIST-like format for fast prototyping and teaching.",
    "provider": "Shanghai Jiao Tong University",
    "region": "International",
    "access": "open",
    "access_note": "pip install medmnist or download from Zenodo (CC BY).",
    "url": "https://medmnist.com/",
    "paper": {
      "label": "Yang J et al. Sci Data 2023;10:41",
      "url": "https://doi.org/10.1038/s41597-022-01721-8"
    },
    "tags": [
      "teaching",
      "benchmark",
      "multi-modality"
    ],
    "category": "imaging"
  },
  {
    "id": "covidx",
    "name": "COVIDx",
    "description": "Open benchmark of tens of thousands of chest X-rays and CT scans assembled from multiple public sources for COVID-19 detection (COVID-Net).",
    "provider": "University of Waterloo / DarwinAI",
    "region": "International",
    "access": "open",
    "access_note": "Download via GitHub instructions / Kaggle.",
    "url": "https://github.com/lindawangg/COVID-Net",
    "paper": {
      "label": "Wang L et al. Sci Rep 2020;10:19549",
      "url": "https://doi.org/10.1038/s41598-020-76550-z"
    },
    "tags": [
      "COVID-19",
      "chest X-ray",
      "CT"
    ],
    "category": "imaging"
  },
  {
    "id": "bimcv-covid19",
    "name": "BIMCV-COVID19+",
    "description": "Chest X-ray and CT studies of COVID-19 patients from the Valencian region with radiological findings, reports and lab results.",
    "provider": "BIMCV, Valencia",
    "region": "Spain",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://bimcv.cipf.es/bimcv-projects/bimcv-covid19/",
    "paper": null,
    "tags": [
      "COVID-19",
      "chest X-ray",
      "CT"
    ],
    "category": "imaging"
  },
  {
    "id": "medpix",
    "name": "MedPix",
    "description": "Free open-access database of ~59,000 medical images and ~12,000 teaching cases indexed by diagnosis, organ system and modality.",
    "provider": "US National Library of Medicine",
    "region": "USA",
    "access": "open",
    "access_note": "Open web access.",
    "url": "https://medpix.nlm.nih.gov/",
    "paper": null,
    "tags": [
      "teaching",
      "cases"
    ],
    "category": "imaging"
  },
  {
    "id": "midrc",
    "name": "MIDRC",
    "full_name": "Medical Imaging and Data Resource Center",
    "description": "Open commons of 500,000+ de-identified imaging studies (chest X-ray, CT, MRI) with clinical data, initially built for COVID-19.",
    "provider": "NIBIB / RSNA / ACR / AAPM",
    "region": "USA",
    "access": "registration",
    "access_note": "Free login and data use agreement in the MIDRC data commons.",
    "url": "https://www.midrc.org/",
    "paper": null,
    "tags": [
      "CT",
      "chest X-ray",
      "COVID-19",
      "hub"
    ],
    "category": "imaging"
  },
  {
    "id": "duke-cspineseg",
    "name": "Duke-CSpineSeg",
    "description": "1,255 cervical spine MRI exams with vertebra and disc segmentations from Duke University Health System.",
    "provider": "Duke University / MIDRC",
    "region": "USA",
    "access": "registration",
    "access_note": "Free MIDRC login.",
    "url": "https://data.midrc.org/discovery/H6K0-A61V",
    "paper": null,
    "tags": [
      "MRI",
      "spine",
      "segmentation"
    ],
    "category": "imaging"
  },
  {
    "id": "duke-breast-mri",
    "name": "Duke Breast Cancer MRI",
    "description": "Pre-operative dynamic contrast-enhanced breast MRI of 922 invasive breast cancer patients with imaging features, pathology and outcomes.",
    "provider": "Duke University via TCIA",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from TCIA.",
    "url": "https://www.cancerimagingarchive.net/collection/duke-breast-cancer-mri/",
    "paper": {
      "label": "Saha A et al. Br J Cancer 2018;119:508",
      "url": "https://doi.org/10.1038/s41416-018-0185-8"
    },
    "tags": [
      "MRI",
      "breast cancer"
    ],
    "category": "imaging"
  },
  {
    "id": "mama-mia",
    "name": "MAMA-MIA",
    "description": "1,506 breast DCE-MRI cases from four TCIA collections with expert tumour segmentations and harmonised clinical data.",
    "provider": "University of Barcelona",
    "region": "International",
    "access": "registration",
    "access_note": "Free Synapse account.",
    "url": "https://www.synapse.org/Synapse:syn60868042/wiki/628716",
    "paper": null,
    "tags": [
      "MRI",
      "breast cancer",
      "segmentation"
    ],
    "category": "imaging"
  },
  {
    "id": "rsna-mammography",
    "name": "RSNA Screening Mammography",
    "description": "Enriched screening mammography set with cancers, biopsy-proven benign findings and negatives, released after the RSNA breast-cancer-detection challenge.",
    "provider": "RSNA",
    "region": "USA / Australia",
    "access": "open",
    "access_note": "Open on AWS Open Data.",
    "url": "https://registry.opendata.aws/rsna-screening-mammography-breast-cancer-detection/",
    "paper": null,
    "tags": [
      "mammography",
      "screening"
    ],
    "category": "imaging"
  },
  {
    "id": "rsna-ich",
    "name": "RSNA Intracranial Hemorrhage Detection",
    "description": "~874,000 head CT slices from 25,000 exams annotated by neuroradiologists for five haemorrhage subtypes.",
    "provider": "RSNA / ASNR",
    "region": "International",
    "access": "registration",
    "access_note": "Free Kaggle account and acceptance of competition rules.",
    "url": "https://www.kaggle.com/competitions/rsna-intracranial-hemorrhage-detection",
    "paper": {
      "label": "Flanders AE et al. Radiol Artif Intell 2020;2:e190211",
      "url": "https://doi.org/10.1148/ryai.2020190211"
    },
    "tags": [
      "head CT",
      "haemorrhage"
    ],
    "category": "imaging"
  },
  {
    "id": "rsna-knee",
    "name": "RSNA Knee Abnormality Detection",
    "description": "Knee MRI studies labelled for twelve abnormal findings from the RSNA challenge.",
    "provider": "RSNA",
    "region": "International",
    "access": "registration",
    "access_note": "Free Kaggle account and acceptance of competition rules.",
    "url": "https://www.kaggle.com/competitions/rsna-knee-abnormality-detection",
    "paper": null,
    "tags": [
      "MRI",
      "knee",
      "challenge"
    ],
    "category": "imaging"
  },
  {
    "id": "ricord",
    "name": "RICORD",
    "full_name": "RSNA International COVID-19 Open Radiology Database",
    "description": "Expert-annotated chest CT (240 studies) and chest radiographs (1,000) of COVID-19 from multiple institutions.",
    "provider": "RSNA via TCIA",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from TCIA.",
    "url": "https://www.rsna.org/covid-19/covid-19-ricord",
    "paper": {
      "label": "Tsai EB et al. Radiology 2021;299:E204",
      "url": "https://doi.org/10.1148/radiol.2021203957"
    },
    "tags": [
      "COVID-19",
      "CT",
      "chest X-ray"
    ],
    "category": "imaging"
  },
  {
    "id": "ct-ich",
    "name": "CT-ICH",
    "description": "82 head CT scans of traumatic brain injury patients with intracranial haemorrhage segmentations.",
    "provider": "Al Hilla Teaching Hospital / PhysioNet",
    "region": "Iraq",
    "access": "registration",
    "access_note": "Free PhysioNet account + signed data use agreement.",
    "url": "https://physionet.org/content/ct-ich/",
    "paper": {
      "label": "Hssayeni MD et al. Data 2020;5:14",
      "url": "https://doi.org/10.3390/data5010014"
    },
    "tags": [
      "head CT",
      "segmentation"
    ],
    "category": "imaging"
  },
  {
    "id": "sinoct",
    "name": "SinoCT",
    "description": "9,000+ non-contrast head CT scans labelled normal or abnormal (~1.3 TB) from Chinese hospitals.",
    "provider": "Stanford AIMI",
    "region": "China",
    "access": "registration",
    "access_note": "Free registration and Stanford research-use agreement.",
    "url": "https://stanfordaimi.azurewebsites.net/datasets/3de8cef8-0626-4c5d-ac47-ed0fed22ac99",
    "paper": null,
    "tags": [
      "head CT"
    ],
    "category": "imaging"
  },
  {
    "id": "deeplesion",
    "name": "NIH DeepLesion",
    "description": "32,000+ annotated lesions on 10,594 CT studies of 4,427 patients, mined from radiologist bookmarks at the NIH Clinical Center.",
    "provider": "NIH Clinical Center",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download (Box).",
    "url": "https://nihcc.app.box.com/v/DeepLesion",
    "paper": {
      "label": "Yan K et al. J Med Imaging 2018;5:036501",
      "url": "https://doi.org/10.1117/1.JMI.5.3.036501"
    },
    "tags": [
      "CT",
      "lesion detection"
    ],
    "category": "imaging"
  },
  {
    "id": "abdomenatlas",
    "name": "AbdomenAtlas",
    "description": "Large-scale multi-organ annotations on thousands of abdominal CT volumes (AbdomenAtlas-8K and later releases) produced with AI-assisted expert review.",
    "provider": "Johns Hopkins University",
    "region": "International",
    "access": "open",
    "access_note": "Direct download via Hugging Face.",
    "url": "https://github.com/MrGiovanni/AbdomenAtlas",
    "paper": {
      "label": "Qu C et al. NeurIPS 2023 Datasets & Benchmarks",
      "url": "https://arxiv.org/abs/2305.09666"
    },
    "tags": [
      "CT",
      "segmentation",
      "abdomen"
    ],
    "category": "imaging"
  },
  {
    "id": "flare-pancancer",
    "name": "MICCAI FLARE Pan-cancer CT",
    "description": "17,000+ CT scans with pan-cancer lesion labels from the FLARE challenge for foundation-model segmentation.",
    "provider": "FLARE challenge",
    "region": "International",
    "access": "registration",
    "access_note": "Hugging Face account and acceptance of terms (gated).",
    "url": "https://huggingface.co/datasets/FLARE-MedFM/PancancerCTSeg",
    "paper": null,
    "tags": [
      "CT",
      "cancer",
      "segmentation"
    ],
    "category": "imaging"
  },
  {
    "id": "rexgroundingct",
    "name": "ReXGroundingCT",
    "description": "Chest CT volumes with free-text findings linked to pixel-level 3D segmentations for grounded report generation.",
    "provider": "Harvard Rajpurkar Lab",
    "region": "International",
    "access": "registration",
    "access_note": "Hugging Face account and acceptance of terms (gated).",
    "url": "https://huggingface.co/datasets/rajpurkarlab/ReXGroundingCT",
    "paper": null,
    "tags": [
      "CT",
      "grounding",
      "vision-language"
    ],
    "category": "imaging"
  },
  {
    "id": "rad-chestct",
    "name": "RAD-ChestCT",
    "description": "Initial public release of 3,630 chest CT volumes from Duke with 83 abnormality labels extracted from reports.",
    "provider": "Duke University",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from Zenodo.",
    "url": "https://zenodo.org/records/6406114",
    "paper": {
      "label": "Draelos RL et al. Med Image Anal 2021;67:101857",
      "url": "https://doi.org/10.1016/j.media.2020.101857"
    },
    "tags": [
      "CT",
      "multi-label"
    ],
    "category": "imaging"
  },
  {
    "id": "mr-rate",
    "name": "MR-RATE",
    "description": "Brain and spine MRI volumes paired with radiology reports for vision-language modelling.",
    "provider": "Forithmus",
    "region": "International",
    "access": "registration",
    "access_note": "Hugging Face account and acceptance of terms (gated).",
    "url": "https://huggingface.co/datasets/Forithmus/MR-RATE",
    "paper": null,
    "tags": [
      "MRI",
      "reports",
      "vision-language"
    ],
    "category": "imaging"
  },
  {
    "id": "cirrmri600",
    "name": "CirrMRI600+",
    "description": "628 T1- and T2-weighted abdominal MRI scans of cirrhotic livers with expert segmentations.",
    "provider": "Northwestern University",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from OSF.",
    "url": "https://osf.io/cuk24/",
    "paper": null,
    "tags": [
      "MRI",
      "liver",
      "cirrhosis"
    ],
    "category": "imaging"
  },
  {
    "id": "brax",
    "name": "BRAX",
    "description": "40,967 chest X-rays from Hospital Israelita Albert Einstein, São Paulo, labelled for 14 findings.",
    "provider": "Hospital Israelita Albert Einstein",
    "region": "Brazil",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + DUA.",
    "url": "https://physionet.org/content/brax/",
    "paper": {
      "label": "Reis EP et al. Sci Data 2022;9:487",
      "url": "https://doi.org/10.1038/s41597-022-01608-8"
    },
    "tags": [
      "chest X-ray",
      "Latin America"
    ],
    "category": "imaging"
  },
  {
    "id": "padchest-gr",
    "name": "PadChest-GR",
    "description": "4,555 bilingual (Spanish/English) chest X-ray studies with sentence-level findings grounded to bounding boxes.",
    "provider": "BIMCV / Microsoft",
    "region": "Spain",
    "access": "application",
    "access_note": "Free; request form and research-use agreement.",
    "url": "https://bimcv.cipf.es/bimcv-projects/padchest-gr/",
    "paper": null,
    "tags": [
      "chest X-ray",
      "grounding"
    ],
    "category": "imaging"
  },
  {
    "id": "chexpert-plus",
    "name": "CheXpert Plus",
    "description": "223,462 chest X-ray–report pairs from Stanford with pathology labels, RadGraph annotations and demographics.",
    "provider": "Stanford AIMI",
    "region": "USA",
    "access": "registration",
    "access_note": "Free registration and Stanford research-use agreement.",
    "url": "https://stanfordaimi.azurewebsites.net/datasets/5158c524-d3ab-4e02-96e9-6ee9efc110a1",
    "paper": {
      "label": "Chambon P et al. arXiv 2024",
      "url": "https://arxiv.org/abs/2405.19538"
    },
    "tags": [
      "chest X-ray",
      "reports"
    ],
    "category": "imaging"
  },
  {
    "id": "echonet-pediatric",
    "name": "EchoNet-Pediatric",
    "description": "7,643 paediatric echocardiogram videos with LV tracings and ejection fraction from Lucile Packard Children's Hospital.",
    "provider": "Stanford AIMI",
    "region": "USA",
    "access": "registration",
    "access_note": "Free registration and Stanford research-use agreement.",
    "url": "https://echonet.github.io/pediatric/",
    "paper": {
      "label": "Reddy CD et al. J Am Soc Echocardiogr 2023;36:482",
      "url": "https://doi.org/10.1016/j.echo.2023.01.015"
    },
    "tags": [
      "echocardiography",
      "paediatrics"
    ],
    "category": "imaging"
  },
  {
    "id": "echonet-lvh",
    "name": "EchoNet-LVH",
    "description": "12,000 parasternal long-axis echo videos with wall-thickness measurements for detecting LV hypertrophy and its causes.",
    "provider": "Stanford AIMI",
    "region": "USA",
    "access": "registration",
    "access_note": "Free registration and Stanford research-use agreement.",
    "url": "https://echonet.github.io/lvh/",
    "paper": {
      "label": "Duffy G et al. JAMA Cardiol 2022;7:386",
      "url": "https://doi.org/10.1001/jamacardio.2021.6059"
    },
    "tags": [
      "echocardiography",
      "hypertrophy"
    ],
    "category": "imaging"
  },
  {
    "id": "mnms",
    "name": "M&Ms Cardiac Segmentation",
    "description": "375 cine cardiac MRI studies from four scanner vendors and six centres with ventricular segmentations, for domain-generalisation research.",
    "provider": "University of Barcelona",
    "region": "Europe",
    "access": "registration",
    "access_note": "Free registration on the challenge site.",
    "url": "https://www.ub.edu/mnms/",
    "paper": {
      "label": "Campello VM et al. IEEE Trans Med Imaging 2021;40:3543",
      "url": "https://doi.org/10.1109/TMI.2021.3090082"
    },
    "tags": [
      "cardiac MRI",
      "segmentation"
    ],
    "category": "imaging"
  },
  {
    "id": "openpocus",
    "name": "OpenPOCUS Lung Ultrasound",
    "description": "Point-of-care lung ultrasound clips (300,000+ frames) from 226 emergency department patients with labels.",
    "provider": "OpenPOCUS",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download via GitHub.",
    "url": "https://github.com/kumarandre/OpenPOCUS",
    "paper": null,
    "tags": [
      "ultrasound",
      "POCUS",
      "ED"
    ],
    "category": "imaging"
  },
  {
    "id": "medshapenet",
    "name": "MedShapeNet",
    "description": "100,000+ 3D medical shapes (organs, bones, vessels, instruments) extracted from public segmentation datasets.",
    "provider": "Institute for AI in Medicine, Essen",
    "region": "International",
    "access": "open",
    "access_note": "Direct download and Python API.",
    "url": "https://medshapenet.ikim.nrw/",
    "paper": {
      "label": "Li J et al. arXiv 2023",
      "url": "https://arxiv.org/abs/2308.16139"
    },
    "tags": [
      "3D shapes",
      "anatomy"
    ],
    "category": "imaging"
  },
  {
    "id": "medpix-2",
    "name": "MedPix 2.0",
    "description": "Structured version of NLM MedPix CT and MRI cases with clinical notes, built for multimodal model training.",
    "provider": "CHILab, University of Palermo",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from GitHub.",
    "url": "https://github.com/CHILab1/MedPix-2.0",
    "paper": null,
    "tags": [
      "CT",
      "MRI",
      "cases"
    ],
    "category": "imaging"
  },
  {
    "id": "nightingale",
    "name": "Nightingale Open Science",
    "description": "Non-profit platform of curated, outcome-linked medical datasets (ECG waveforms, biopsy slides, chest X-rays) from health systems worldwide.",
    "provider": "Nightingale Open Science",
    "region": "International",
    "access": "application",
    "access_note": "Free; register and apply; analysis in a secure cloud.",
    "url": "https://docs.nightingalescience.org/",
    "paper": null,
    "tags": [
      "outcomes",
      "hub"
    ],
    "category": "imaging"
  },
  {
    "id": "camelyon",
    "name": "CAMELYON16/17",
    "description": "Whole-slide images of sentinel lymph nodes from breast-cancer patients with pixel-level metastasis annotations (1,399 WSIs) from Dutch medical centres.",
    "provider": "Radboud UMC / UMC Utrecht",
    "region": "Netherlands",
    "access": "open",
    "access_note": "Open download (CC0) via AWS Open Data / GigaScience.",
    "url": "https://camelyon17.grand-challenge.org/Data/",
    "paper": {
      "label": "Ehteshami Bejnordi B et al. JAMA 2017;318:2199",
      "url": "https://doi.org/10.1001/jama.2017.14585"
    },
    "tags": [
      "histopathology",
      "WSI",
      "breast cancer"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "panda",
    "name": "PANDA",
    "full_name": "Prostate cANcer graDe Assessment Challenge",
    "description": "~11,000 prostate biopsy whole-slide images from Radboud and Karolinska with ISUP grade and Gleason labels.",
    "provider": "Radboud UMC / Karolinska Institutet",
    "region": "Netherlands / Sweden",
    "access": "registration",
    "access_note": "Free Kaggle account and acceptance of competition rules.",
    "url": "https://www.kaggle.com/c/prostate-cancer-grade-assessment",
    "paper": {
      "label": "Bulten W et al. Nat Med 2022;28:154",
      "url": "https://doi.org/10.1038/s41591-021-01620-2"
    },
    "tags": [
      "histopathology",
      "prostate",
      "Gleason"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "pcam",
    "name": "PatchCamelyon (PCam)",
    "description": "327,680 96×96 histopathology patches derived from CAMELYON16 with binary metastasis labels — an MNIST-sized pathology benchmark.",
    "provider": "Veeling et al.",
    "region": "Netherlands",
    "access": "open",
    "access_note": "Direct download (CC0).",
    "url": "https://github.com/basveeling/pcam",
    "paper": {
      "label": "Veeling BS et al. MICCAI 2018",
      "url": "https://doi.org/10.1007/978-3-030-00934-2_24"
    },
    "tags": [
      "histopathology",
      "patches",
      "benchmark"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "breakhis",
    "name": "BreakHis",
    "description": "7,909 microscopic images of breast tumour tissue from 82 patients at four magnifications, labelled benign or malignant with subtypes.",
    "provider": "Federal University of Paraná",
    "region": "Brazil",
    "access": "open",
    "access_note": "Direct download for research use.",
    "url": "https://web.inf.ufpr.br/vri/databases/breast-cancer-histopathological-database-breakhis/",
    "paper": {
      "label": "Spanhol FA et al. IEEE Trans Biomed Eng 2016;63:1455",
      "url": "https://doi.org/10.1109/TBME.2015.2496264"
    },
    "tags": [
      "histopathology",
      "breast cancer"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "isic",
    "name": "ISIC Archive",
    "description": "The largest public collection of dermoscopic images of skin lesions (>400,000), with diagnoses and metadata, home of the annual ISIC melanoma challenges.",
    "provider": "International Skin Imaging Collaboration",
    "region": "International",
    "access": "open",
    "access_note": "Open download and API (mostly CC BY / CC BY-NC).",
    "url": "https://www.isic-archive.com/",
    "paper": null,
    "tags": [
      "dermoscopy",
      "melanoma"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "ham10000",
    "name": "HAM10000",
    "description": "10,015 dermoscopic images of common pigmented skin lesions across 7 diagnostic categories, >50% histopathology-confirmed.",
    "provider": "Medical University of Vienna",
    "region": "Austria / Australia",
    "access": "open",
    "access_note": "Direct download from Harvard Dataverse (CC BY-NC).",
    "url": "https://doi.org/10.7910/DVN/DBW86T",
    "paper": {
      "label": "Tschandl P et al. Sci Data 2018;5:180161",
      "url": "https://doi.org/10.1038/sdata.2018.161"
    },
    "tags": [
      "dermoscopy",
      "skin lesions"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "fitzpatrick17k",
    "name": "Fitzpatrick 17k",
    "description": "16,577 clinical skin-condition images across 114 conditions annotated with Fitzpatrick skin type to study fairness in dermatology AI.",
    "provider": "MIT / Groh et al.",
    "region": "International",
    "access": "open",
    "access_note": "Image URLs and labels on GitHub.",
    "url": "https://github.com/mattgroh/fitzpatrick17k",
    "paper": {
      "label": "Groh M et al. CVPR Workshops 2021",
      "url": "https://arxiv.org/abs/2104.09957"
    },
    "tags": [
      "dermatology",
      "skin of colour",
      "fairness"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "ddi",
    "name": "DDI",
    "full_name": "Diverse Dermatology Images",
    "description": "656 biopsy-proven clinical images emphasising Fitzpatrick skin types V–VI and uncommon diseases, built to expose bias in dermatology algorithms.",
    "provider": "Stanford AIMI",
    "region": "USA",
    "access": "registration",
    "access_note": "Free registration and research-use agreement (Stanford AIMI).",
    "url": "https://ddi-dataset.github.io/",
    "paper": {
      "label": "Daneshjou R et al. Sci Adv 2022;8:eabq6147",
      "url": "https://doi.org/10.1126/sciadv.abq6147"
    },
    "tags": [
      "dermatology",
      "skin of colour",
      "fairness"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "pad-ufes-20",
    "name": "PAD-UFES-20",
    "description": "2,298 smartphone images of skin lesions from 1,373 Brazilian patients with clinical metadata across six diagnoses.",
    "provider": "Federal University of Espírito Santo",
    "region": "Brazil",
    "access": "open",
    "access_note": "Direct download from Mendeley Data (CC BY 4.0).",
    "url": "https://data.mendeley.com/datasets/zr7vgbcyr2/1",
    "paper": {
      "label": "Pacheco AGC et al. Data Brief 2020;32:106221",
      "url": "https://doi.org/10.1016/j.dib.2020.106221"
    },
    "tags": [
      "dermatology",
      "smartphone",
      "LMIC"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "eyepacs-kaggle",
    "name": "EyePACS Diabetic Retinopathy (Kaggle)",
    "description": "~88,000 retinal fundus photographs graded 0–4 for diabetic retinopathy, released for the 2015 Kaggle competition.",
    "provider": "EyePACS / California Healthcare Foundation",
    "region": "USA",
    "access": "registration",
    "access_note": "Free Kaggle account and acceptance of competition rules.",
    "url": "https://www.kaggle.com/c/diabetic-retinopathy-detection",
    "paper": null,
    "tags": [
      "fundus",
      "diabetic retinopathy"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "aptos-2019",
    "name": "APTOS 2019 Blindness Detection",
    "description": "~5,600 fundus images from rural Indian screening programmes (Aravind Eye Hospital) graded for diabetic retinopathy severity.",
    "provider": "Asia Pacific Tele-Ophthalmology Society",
    "region": "India",
    "access": "registration",
    "access_note": "Free Kaggle account and acceptance of competition rules.",
    "url": "https://www.kaggle.com/c/aptos2019-blindness-detection",
    "paper": null,
    "tags": [
      "fundus",
      "diabetic retinopathy",
      "LMIC"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "idrid",
    "name": "IDRiD",
    "full_name": "Indian Diabetic Retinopathy Image Dataset",
    "description": "516 high-resolution fundus images with pixel-level lesion annotations, DR/DME grades and optic disc/fovea locations.",
    "provider": "SGGS Institute, Nanded",
    "region": "India",
    "access": "registration",
    "access_note": "Free IEEE DataPort account (open-access dataset).",
    "url": "https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid",
    "paper": {
      "label": "Porwal P et al. Data 2018;3:25",
      "url": "https://doi.org/10.3390/data3030025"
    },
    "tags": [
      "fundus",
      "segmentation",
      "LMIC"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "messidor-2",
    "name": "Messidor-2",
    "description": "1,748 macula-centred fundus images from 874 diabetic examinations, widely used to validate DR screening algorithms.",
    "provider": "LaTIM / ADCIS",
    "region": "France",
    "access": "registration",
    "access_note": "Free; request via licence form.",
    "url": "https://www.adcis.net/en/third-party/messidor2/",
    "paper": {
      "label": "Decencière E et al. Image Anal Stereol 2014;33:231",
      "url": "https://doi.org/10.5566/ias.1155"
    },
    "tags": [
      "fundus",
      "diabetic retinopathy"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "drive",
    "name": "DRIVE",
    "full_name": "Digital Retinal Images for Vessel Extraction",
    "description": "40 colour fundus images with manual retinal-vessel segmentations — the classic vessel-segmentation benchmark.",
    "provider": "Image Sciences Institute, Utrecht",
    "region": "Netherlands",
    "access": "registration",
    "access_note": "Free Grand Challenge account.",
    "url": "https://drive.grand-challenge.org/",
    "paper": {
      "label": "Staal J et al. IEEE Trans Med Imaging 2004;23:501",
      "url": "https://doi.org/10.1109/TMI.2004.825627"
    },
    "tags": [
      "fundus",
      "vessel segmentation"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "refuge",
    "name": "REFUGE",
    "full_name": "Retinal Fundus Glaucoma Challenge",
    "description": "1,200 fundus images with glaucoma labels and optic disc/cup segmentations from Chinese hospitals.",
    "provider": "Baidu / REFUGE consortium",
    "region": "China",
    "access": "registration",
    "access_note": "Free Grand Challenge account.",
    "url": "https://refuge.grand-challenge.org/",
    "paper": {
      "label": "Orlando JI et al. Med Image Anal 2020;59:101570",
      "url": "https://doi.org/10.1016/j.media.2019.101570"
    },
    "tags": [
      "fundus",
      "glaucoma"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "nct-crc-he-100k",
    "name": "NCT-CRC-HE-100K",
    "description": "100,000 H&E image patches of human colorectal cancer and normal tissue in nine tissue classes.",
    "provider": "NCT Heidelberg / Kather lab",
    "region": "Germany",
    "access": "open",
    "access_note": "Direct download from Zenodo (CC BY 4.0).",
    "url": "https://zenodo.org/records/1214456",
    "paper": {
      "label": "Kather JN et al. PLoS Med 2019;16:e1002730",
      "url": "https://doi.org/10.1371/journal.pmed.1002730"
    },
    "tags": [
      "histopathology",
      "colorectal cancer"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "starc-9",
    "name": "STARC-9",
    "description": "630,000 colorectal tissue image tiles across nine tissue classes for training pathology classifiers.",
    "provider": "Path2AI",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from Hugging Face.",
    "url": "https://huggingface.co/datasets/Path2AI/STARC-9",
    "paper": null,
    "tags": [
      "histopathology",
      "colorectal cancer"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "pannuke",
    "name": "PanNuke",
    "description": "Nuclei instance segmentation and classification (~200,000 nuclei) across 19 tissue types.",
    "provider": "University of Warwick",
    "region": "United Kingdom",
    "access": "open",
    "access_note": "Direct download (CC BY-NC-SA).",
    "url": "https://warwick.ac.uk/fac/cross_fac/tia/data/pannuke",
    "paper": {
      "label": "Gamper J et al. arXiv 2020",
      "url": "https://arxiv.org/abs/2003.10778"
    },
    "tags": [
      "histopathology",
      "nuclei"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "nucls",
    "name": "NuCLS",
    "description": "220,000+ crowdsourced and expert-reviewed nucleus labels from TCGA breast cancer slides.",
    "provider": "Emory University / Northwestern",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download (CC0).",
    "url": "https://sites.google.com/view/nucls",
    "paper": {
      "label": "Amgad M et al. GigaScience 2022;11:giac037",
      "url": "https://doi.org/10.1093/gigascience/giac037"
    },
    "tags": [
      "histopathology",
      "nuclei",
      "breast cancer"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "diagset",
    "name": "DiagSet",
    "description": "Prostate cancer histopathology: 2.6 million patches and 4,675 whole-slide scans with diagnoses.",
    "provider": "AGH University of Kraków",
    "region": "Poland",
    "access": "open",
    "access_note": "Download links on GitHub.",
    "url": "https://github.com/michalkoziarski/DiagSet",
    "paper": {
      "label": "Koziarski M et al. Sci Rep 2024;14:6780",
      "url": "https://doi.org/10.1038/s41598-024-52183-4"
    },
    "tags": [
      "histopathology",
      "prostate cancer"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "plism",
    "name": "PLISM",
    "description": "Pathology images of the same tissues captured with different whole-slide scanners and smartphones, for studying domain shift.",
    "provider": "Ochi M et al.",
    "region": "Japan",
    "access": "open",
    "access_note": "Direct download from figshare.",
    "url": "https://plus.figshare.com/collections/Pathology_Images_of_Scanners_and_Mobilephones_PLISM_Dataset/6773925",
    "paper": null,
    "tags": [
      "histopathology",
      "domain shift"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "3d-pathology-triage",
    "name": "3D Pathology Triage Datasets",
    "description": "3D pathology image volumes of prostate and oesophageal tissue with binary labels for deep-learning triage of 3D microscopy.",
    "provider": "University of Washington",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from Zenodo.",
    "url": "https://zenodo.org/records/20052262",
    "paper": null,
    "tags": [
      "3D pathology",
      "microscopy"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "nlm-malaria",
    "name": "NLM Malaria Cell Images",
    "description": "27,558 segmented thin blood-smear red-cell images, evenly split between Plasmodium-parasitised and uninfected.",
    "provider": "US National Library of Medicine",
    "region": "Bangladesh / USA",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://lhncbc.nlm.nih.gov/LHC-research/LHC-projects/image-processing/malaria-datasheet.html",
    "paper": {
      "label": "Rajaraman S et al. PeerJ 2018;6:e4568",
      "url": "https://doi.org/10.7717/peerj.4568"
    },
    "tags": [
      "malaria",
      "microscopy",
      "LMIC"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "ph2",
    "name": "PH²",
    "description": "200 dermoscopic images of melanocytic lesions with expert segmentations and diagnoses from Hospital Pedro Hispano, Portugal.",
    "provider": "University of Porto",
    "region": "Portugal",
    "access": "registration",
    "access_note": "Free; fill a short registration form to download.",
    "url": "https://www.fc.up.pt/addi/ph2%20database.html",
    "paper": {
      "label": "Mendonça T et al. IEEE EMBC 2013",
      "url": "https://doi.org/10.1109/EMBC.2013.6610779"
    },
    "tags": [
      "dermoscopy",
      "melanoma"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "med-node",
    "name": "MED-NODE",
    "description": "170 non-dermoscopic clinical images of melanoma and naevi from the University Medical Center Groningen.",
    "provider": "University of Groningen",
    "region": "Netherlands",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://www.cs.rug.nl/~imaging/databases/melanoma_naevi/",
    "paper": {
      "label": "Giotis I et al. Expert Syst Appl 2015;42:6578",
      "url": "https://doi.org/10.1016/j.eswa.2015.04.034"
    },
    "tags": [
      "dermatology",
      "melanoma"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "derm7pt",
    "name": "Derm7pt (Seven-Point Checklist)",
    "description": "1,011 cases with clinical and dermoscopic images annotated with the seven-point melanoma checklist and diagnoses.",
    "provider": "Simon Fraser University",
    "region": "Canada / Italy",
    "access": "registration",
    "access_note": "Free; request via online form.",
    "url": "http://derm.cs.sfu.ca/",
    "paper": {
      "label": "Kawahara J et al. IEEE J Biomed Health Inform 2019;23:538",
      "url": "https://doi.org/10.1109/JBHI.2018.2824327"
    },
    "tags": [
      "dermoscopy",
      "melanoma"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "derm-ddx",
    "name": "Dermatology DDx (Uncertain Ground Truth)",
    "description": "1,947 teledermatology cases with differential diagnoses from multiple dermatologists across 419 conditions, for studying label uncertainty.",
    "provider": "Google DeepMind",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from GitHub.",
    "url": "https://github.com/google-deepmind/uncertain_ground_truth",
    "paper": null,
    "tags": [
      "dermatology",
      "label uncertainty"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "snu-derm",
    "name": "SNU Dermatology Dataset",
    "description": "2,201 clinical skin images across 174 diseases with quizzes, released with the Seoul National University dermatology AI studies.",
    "provider": "Seoul National University",
    "region": "South Korea",
    "access": "open",
    "access_note": "Direct download from figshare.",
    "url": "https://figshare.com/articles/dataset/SNU_SNU_MELANOMA_and_Reddit_dataset_Quiz/6454973",
    "paper": null,
    "tags": [
      "dermatology",
      "Asia"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "encode-skin-color",
    "name": "ENCoDE Skin Colour Dataset",
    "description": "Objective skin-tone measurements linked to EHR data (167 variables) to study pulse-oximetry and other device bias.",
    "provider": "PhysioNet",
    "region": "USA",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + DUA.",
    "url": "https://physionet.org/content/encode-skin-color/",
    "paper": null,
    "tags": [
      "skin tone",
      "bias",
      "pulse oximetry"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "brset",
    "name": "BRSET",
    "full_name": "Brazilian Multilabel Ophthalmological Dataset",
    "description": "16,266 retinal fundus photographs from 8,524 Brazilian patients with demographics and multi-label diagnoses.",
    "provider": "Federal University of São Paulo / MIT",
    "region": "Brazil",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + DUA.",
    "url": "https://physionet.org/content/brazilian-ophthalmological/",
    "paper": {
      "label": "Nakayama LF et al. PLOS Digit Health 2024;3:e0000454",
      "url": "https://doi.org/10.1371/journal.pdig.0000454"
    },
    "tags": [
      "fundus",
      "Latin America"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "mbrset",
    "name": "mBRSET",
    "description": "5,164 smartphone retinal images from 1,291 diabetic patients in Brazil with DR grades and systemic data.",
    "provider": "Federal University of São Paulo / MIT",
    "region": "Brazil",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + DUA.",
    "url": "https://physionet.org/content/mbrset/",
    "paper": null,
    "tags": [
      "fundus",
      "smartphone",
      "diabetic retinopathy"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "hygd",
    "name": "Hillel Yaffe Glaucoma Dataset (HYGD)",
    "description": "747 fundus images with gold-standard glaucoma labels based on full ophthalmic examination.",
    "provider": "Hillel Yaffe Medical Center",
    "region": "Israel",
    "access": "open",
    "access_note": "Direct download from PhysioNet.",
    "url": "https://physionet.org/content/hillel-yaffe-glaucoma-dataset/",
    "paper": null,
    "tags": [
      "fundus",
      "glaucoma"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "ffa-ir",
    "name": "FFA-IR",
    "description": "47,247 fundus fluorescein angiography images from 10,790 reports with lesion annotations, for explainable report generation.",
    "provider": "Sun Yat-sen University",
    "region": "China",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + DUA.",
    "url": "https://physionet.org/content/ffa-ir-medical-report/",
    "paper": null,
    "tags": [
      "angiography",
      "reports"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "hrf",
    "name": "HRF",
    "full_name": "High-Resolution Fundus Image Database",
    "description": "45 high-resolution fundus images (healthy, diabetic retinopathy, glaucoma) with vessel segmentations.",
    "provider": "Friedrich-Alexander-Universität Erlangen",
    "region": "Germany",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://www5.cs.fau.de/research/data/fundus-images/",
    "paper": {
      "label": "Budai A et al. Int J Biomed Imaging 2013:154860",
      "url": "https://doi.org/10.1155/2013/154860"
    },
    "tags": [
      "fundus",
      "vessel segmentation"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "riga",
    "name": "RIGA",
    "description": "750 fundus images with optic disc and cup boundaries marked by six ophthalmologists (4,500 annotations) for glaucoma research.",
    "provider": "University of Michigan / Riyadh hospitals",
    "region": "Saudi Arabia / France",
    "access": "open",
    "access_note": "Direct download from Deep Blue Data.",
    "url": "https://deepblue.lib.umich.edu/data/concern/data_sets/3b591905z",
    "paper": {
      "label": "Almazroa A et al. Proc SPIE 2018;10579",
      "url": "https://doi.org/10.1117/12.2293584"
    },
    "tags": [
      "fundus",
      "glaucoma"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "dr-hagis",
    "name": "DR HAGIS",
    "description": "39 fundus images from a UK diabetic screening programme with manual vessel segmentations, covering comorbidities.",
    "provider": "University of Manchester",
    "region": "United Kingdom",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://personalpages.manchester.ac.uk/staff/niall.p.mcloughlin/",
    "paper": {
      "label": "Holm S et al. J Med Imaging 2017;4:014503",
      "url": "https://doi.org/10.1117/1.JMI.4.1.014503"
    },
    "tags": [
      "fundus",
      "vessel segmentation"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "sustech-sysu",
    "name": "SUSTech-SYSU Corneal Ulcer Dataset",
    "description": "712 fluorescein-stained ocular surface images with corneal ulcer segmentations and type/grade labels.",
    "provider": "Southern University of Science and Technology / Sun Yat-sen University",
    "region": "China",
    "access": "open",
    "access_note": "Direct download from figshare.",
    "url": "https://figshare.com/articles/dataset/Metadata_record_for_The_SUSTech-SYSU_dataset_for_automatically_segmenting_and_classifying_corneal_ulcers/11371803/2",
    "paper": {
      "label": "Deng L et al. Sci Data 2020;7:23",
      "url": "https://doi.org/10.1038/s41597-020-0360-7"
    },
    "tags": [
      "cornea",
      "ulcer",
      "segmentation"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "rim-one",
    "name": "RIM-ONE DL",
    "description": "Fundus images with optic disc and cup segmentations, curated for deep-learning glaucoma assessment.",
    "provider": "Universidad de La Laguna",
    "region": "Spain",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://github.com/miag-ull/rim-one-dl",
    "paper": null,
    "tags": [
      "fundus",
      "glaucoma"
    ],
    "category": "pathology-derm-eye"
  },
  {
    "id": "adni",
    "name": "ADNI",
    "full_name": "Alzheimer's Disease Neuroimaging Initiative",
    "description": "Longitudinal MRI, PET, CSF/blood biomarkers, genetics and cognitive data from >2,000 participants (normal, MCI, Alzheimer's) since 2004.",
    "provider": "ADNI / LONI, University of Southern California",
    "region": "USA / Canada",
    "access": "application",
    "access_note": "Free; online application via LONI IDA and Data Use Agreement.",
    "url": "https://adni.loni.usc.edu/data-samples/adni-data/",
    "paper": {
      "label": "Petersen RC et al. Neurology 2010;74:201",
      "url": "https://doi.org/10.1212/WNL.0b013e3181cb3e25"
    },
    "tags": [
      "MRI",
      "PET",
      "dementia"
    ],
    "category": "neuro"
  },
  {
    "id": "oasis",
    "name": "OASIS",
    "full_name": "Open Access Series of Imaging Studies",
    "description": "Brain MRI (and PET in OASIS-3/4) with clinical and cognitive data across normal ageing and Alzheimer's disease, from Washington University's Knight ADRC.",
    "provider": "Washington University in St. Louis",
    "region": "USA",
    "access": "registration",
    "access_note": "Free; agree to the Data Use Agreement and register on NITRC-IR/XNAT Central.",
    "url": "https://sites.wustl.edu/oasisbrains/",
    "paper": {
      "label": "Marcus DS et al. J Cogn Neurosci 2007;19:1498",
      "url": "https://doi.org/10.1162/jocn.2007.19.9.1498"
    },
    "tags": [
      "MRI",
      "ageing",
      "dementia"
    ],
    "category": "neuro"
  },
  {
    "id": "openneuro",
    "name": "OpenNeuro",
    "description": "Open platform hosting 1,000+ BIDS-formatted neuroimaging datasets (MRI, MEG, EEG, iEEG, PET) contributed by researchers worldwide.",
    "provider": "Stanford Center for Reproducible Neuroscience",
    "region": "International",
    "access": "open",
    "access_note": "Direct download (mostly CC0).",
    "url": "https://openneuro.org/",
    "paper": {
      "label": "Markiewicz CJ et al. eLife 2021;10:e71774",
      "url": "https://doi.org/10.7554/eLife.71774"
    },
    "tags": [
      "MRI",
      "EEG",
      "BIDS",
      "hub"
    ],
    "category": "neuro"
  },
  {
    "id": "hcp",
    "name": "Human Connectome Project",
    "description": "High-quality multimodal MRI (structural, resting/task fMRI, diffusion) and behavioural data from ~1,200 healthy young adults, plus Lifespan and disease extensions.",
    "provider": "WU-Minn HCP Consortium",
    "region": "USA",
    "access": "registration",
    "access_note": "Free ConnectomeDB account + Open Access Data Use Terms; restricted data by application.",
    "url": "https://www.humanconnectome.org/study/hcp-young-adult",
    "paper": {
      "label": "Van Essen DC et al. NeuroImage 2013;80:62",
      "url": "https://doi.org/10.1016/j.neuroimage.2013.05.041"
    },
    "tags": [
      "MRI",
      "fMRI",
      "connectome"
    ],
    "category": "neuro"
  },
  {
    "id": "abide",
    "name": "ABIDE",
    "full_name": "Autism Brain Imaging Data Exchange",
    "description": "Resting-state fMRI and structural MRI from >2,000 individuals with autism and typical controls aggregated from 24+ international sites.",
    "provider": "International Neuroimaging Data-sharing Initiative",
    "region": "International",
    "access": "registration",
    "access_note": "Free NITRC registration.",
    "url": "https://fcon_1000.projects.nitrc.org/indi/abide/",
    "paper": {
      "label": "Di Martino A et al. Mol Psychiatry 2014;19:659",
      "url": "https://doi.org/10.1038/mp.2013.78"
    },
    "tags": [
      "fMRI",
      "autism"
    ],
    "category": "neuro"
  },
  {
    "id": "abcd",
    "name": "ABCD Study",
    "full_name": "Adolescent Brain Cognitive Development Study",
    "description": "Largest long-term study of brain development in the US: ~11,800 children followed from age 9–10 with MRI, cognition, mental health, substance use and genetics.",
    "provider": "NIH / NBDC",
    "region": "USA",
    "access": "application",
    "access_note": "Free; institutional Data Use Certification via the NIH Brain Development Cohorts data hub.",
    "url": "https://abcdstudy.org/",
    "paper": {
      "label": "Casey BJ et al. Dev Cogn Neurosci 2018;32:43",
      "url": "https://doi.org/10.1016/j.dcn.2018.03.001"
    },
    "tags": [
      "MRI",
      "child development",
      "mental health"
    ],
    "category": "neuro"
  },
  {
    "id": "nda",
    "name": "NIMH Data Archive (NDA)",
    "description": "Federated archive of human-subjects mental-health research data (clinical, imaging, genomics) from hundreds of NIH-funded studies, including ABCD and HCP lifespan.",
    "provider": "US National Institute of Mental Health",
    "region": "USA",
    "access": "application",
    "access_note": "Free; Data Use Certification signed by your institution.",
    "url": "https://nda.nih.gov/",
    "paper": null,
    "tags": [
      "psychiatry",
      "hub"
    ],
    "category": "neuro"
  },
  {
    "id": "ixi",
    "name": "IXI Dataset",
    "description": "~600 MR images (T1, T2, PD, MRA, DTI) from healthy subjects scanned at three London hospitals.",
    "provider": "Imperial College London",
    "region": "United Kingdom",
    "access": "open",
    "access_note": "Direct download (CC BY-SA 3.0).",
    "url": "https://brain-development.org/ixi-dataset/",
    "paper": null,
    "tags": [
      "MRI",
      "healthy controls"
    ],
    "category": "neuro"
  },
  {
    "id": "brats",
    "name": "BraTS",
    "full_name": "Brain Tumor Segmentation Challenge",
    "description": "Multi-institutional multiparametric MRI (T1, T1-Gd, T2, FLAIR) of gliomas and other brain tumours with expert segmentations of tumour sub-regions.",
    "provider": "CBICA, University of Pennsylvania / RSNA-ASNR-MICCAI",
    "region": "International",
    "access": "registration",
    "access_note": "Free Synapse registration for the yearly challenge data.",
    "url": "https://www.synapse.org/brats",
    "paper": {
      "label": "Menze BH et al. IEEE Trans Med Imaging 2015;34:1993",
      "url": "https://doi.org/10.1109/TMI.2014.2377694"
    },
    "tags": [
      "MRI",
      "brain tumour",
      "segmentation"
    ],
    "category": "neuro"
  },
  {
    "id": "tuh-eeg",
    "name": "TUH EEG Corpus",
    "description": "The world's largest open clinical EEG corpus (>30,000 recordings) from Temple University Hospital with physician reports; subsets annotated for seizures, artefacts and abnormality.",
    "provider": "Neural Engineering Data Consortium, Temple University",
    "region": "USA",
    "access": "registration",
    "access_note": "Free; submit a signed data-sharing form to receive credentials.",
    "url": "https://isip.piconepress.com/projects/nedc/html/tuh_eeg/",
    "paper": {
      "label": "Obeid I, Picone J. Front Neurosci 2016;10:196",
      "url": "https://doi.org/10.3389/fnins.2016.00196"
    },
    "tags": [
      "EEG",
      "seizures"
    ],
    "category": "neuro"
  },
  {
    "id": "chb-mit",
    "name": "CHB-MIT Scalp EEG Database",
    "description": "Long-term scalp EEG of 22 paediatric subjects with intractable seizures at Boston Children's Hospital, with annotated seizure onsets.",
    "provider": "Boston Children's Hospital / MIT via PhysioNet",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download (ODC-By).",
    "url": "https://physionet.org/content/chbmit/",
    "paper": null,
    "tags": [
      "EEG",
      "epilepsy",
      "paediatrics"
    ],
    "category": "neuro"
  },
  {
    "id": "siena-eeg",
    "name": "Siena Scalp EEG Database",
    "description": "EEG recordings of 14 adult epilepsy patients from the University of Siena with seizure annotations by neurologists.",
    "provider": "University of Siena via PhysioNet",
    "region": "Italy",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://physionet.org/content/siena-scalp-eeg/",
    "paper": {
      "label": "Detti P et al. Processes 2020;8:846",
      "url": "https://doi.org/10.3390/pr8070846"
    },
    "tags": [
      "EEG",
      "epilepsy"
    ],
    "category": "neuro"
  },
  {
    "id": "eegmmidb",
    "name": "EEG Motor Movement/Imagery Dataset",
    "description": "1,500+ 64-channel EEG recordings from 109 volunteers performing real and imagined motor tasks; widely used for brain–computer interface research.",
    "provider": "BCI2000 / PhysioNet",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://physionet.org/content/eegmmidb/",
    "paper": {
      "label": "Schalk G et al. IEEE Trans Biomed Eng 2004;51:1034",
      "url": "https://doi.org/10.1109/TBME.2004.827072"
    },
    "tags": [
      "EEG",
      "BCI"
    ],
    "category": "neuro"
  },
  {
    "id": "neurodatasets-r",
    "name": "NeuroDataSets (R package)",
    "description": "R package bundling a diverse collection of ready-to-use datasets on the brain, nervous system and neurological and psychiatric disorders, each documented with its original source and licence.",
    "provider": "Renzo Caceres Rossi",
    "region": "International",
    "access": "open",
    "access_note": "install.packages(\"NeuroDataSets\") from CRAN; source on GitHub.",
    "url": "https://lightbluetitan.github.io/neurodatasets/",
    "paper": null,
    "tags": [
      "R package",
      "teaching",
      "hub"
    ],
    "category": "neuro"
  },
  {
    "id": "upenn-gbm",
    "name": "UPENN-GBM",
    "description": "Multiparametric MRI of 630 glioblastoma patients with tumour segmentations, radiomic features, genomics and survival.",
    "provider": "University of Pennsylvania via TCIA",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from TCIA.",
    "url": "https://www.cancerimagingarchive.net/collection/upenn-gbm/",
    "paper": {
      "label": "Bakas S et al. Sci Data 2022;9:453",
      "url": "https://doi.org/10.1038/s41597-022-01560-7"
    },
    "tags": [
      "MRI",
      "glioblastoma"
    ],
    "category": "neuro"
  },
  {
    "id": "brats-africa",
    "name": "BraTS-Africa",
    "description": "Pre-operative brain MRI with glioma segmentations for 146 patients from Nigeria, addressing under-representation of African populations.",
    "provider": "BraTS / Medical Artificial Intelligence Lab, Lagos",
    "region": "Nigeria",
    "access": "open",
    "access_note": "Direct download from TCIA.",
    "url": "https://www.cancerimagingarchive.net/collection/brats-africa/",
    "paper": {
      "label": "Adewole M et al. arXiv 2023",
      "url": "https://arxiv.org/abs/2305.19369"
    },
    "tags": [
      "MRI",
      "glioma",
      "Africa"
    ],
    "category": "neuro"
  },
  {
    "id": "atlas-stroke",
    "name": "ATLAS v2.0",
    "full_name": "Anatomical Tracings of Lesions After Stroke",
    "description": "955 T1-weighted MRIs of stroke patients with manually segmented lesions from 33 cohorts worldwide.",
    "provider": "University of Southern California",
    "region": "International",
    "access": "registration",
    "access_note": "Free ICPSR login and data use agreement.",
    "url": "https://www.icpsr.umich.edu/web/ADDEP/studies/36684",
    "paper": {
      "label": "Liew SL et al. Sci Data 2022;9:320",
      "url": "https://doi.org/10.1038/s41597-022-01401-7"
    },
    "tags": [
      "MRI",
      "stroke",
      "segmentation"
    ],
    "category": "neuro"
  },
  {
    "id": "nyumets",
    "name": "NYUMets: Brain",
    "description": "Longitudinal MRI, segmentations, treatment (gamma knife) and outcome data of patients with brain metastases from NYU Langone.",
    "provider": "NYU Langone Health",
    "region": "USA",
    "access": "application",
    "access_note": "Free; request access and sign a data use agreement.",
    "url": "https://nyumets.org/",
    "paper": null,
    "tags": [
      "MRI",
      "brain metastases",
      "longitudinal"
    ],
    "category": "neuro"
  },
  {
    "id": "eegeyenet",
    "name": "EEGEyeNet",
    "description": "Simultaneous high-density EEG and eye-tracking recordings from 356 participants (47+ hours) for gaze estimation benchmarks.",
    "provider": "ETH Zürich / University of Zürich",
    "region": "Switzerland",
    "access": "open",
    "access_note": "Direct download from OSF.",
    "url": "https://osf.io/ktv7m/",
    "paper": {
      "label": "Kastrati A et al. NeurIPS 2021 Datasets & Benchmarks",
      "url": "https://arxiv.org/abs/2111.05100"
    },
    "tags": [
      "EEG",
      "eye tracking"
    ],
    "category": "neuro"
  },
  {
    "id": "mitdb",
    "name": "MIT-BIH Arrhythmia Database",
    "description": "48 half-hour two-lead ambulatory ECG recordings with beat-by-beat cardiologist annotations — the reference standard for arrhythmia-detection algorithms since 1980.",
    "provider": "MIT / Beth Israel Hospital via PhysioNet",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://physionet.org/content/mitdb/",
    "paper": {
      "label": "Moody GB, Mark RG. IEEE Eng Med Biol Mag 2001;20:45",
      "url": "https://doi.org/10.1109/51.932724"
    },
    "tags": [
      "ECG",
      "arrhythmia",
      "benchmark"
    ],
    "category": "signals"
  },
  {
    "id": "ptb-xl",
    "name": "PTB-XL",
    "description": "21,799 clinical 12-lead, 10-second ECGs from 18,869 patients with SCP-ECG diagnostic, form and rhythm statements annotated by cardiologists.",
    "provider": "Physikalisch-Technische Bundesanstalt via PhysioNet",
    "region": "Germany",
    "access": "open",
    "access_note": "Direct download (CC BY 4.0).",
    "url": "https://physionet.org/content/ptb-xl/",
    "paper": {
      "label": "Wagner P et al. Sci Data 2020;7:154",
      "url": "https://doi.org/10.1038/s41597-020-0495-6"
    },
    "tags": [
      "ECG",
      "12-lead"
    ],
    "category": "signals"
  },
  {
    "id": "mimic-iv-ecg",
    "name": "MIMIC-IV-ECG",
    "description": "~800,000 diagnostic 12-lead ECGs from ~160,000 patients linked to MIMIC-IV, with machine measurements and cardiologist report text.",
    "provider": "MIT Laboratory for Computational Physiology",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from PhysioNet (ODbL); linkage to clinical data needs MIMIC-IV credentialed access.",
    "url": "https://physionet.org/content/mimic-iv-ecg/",
    "paper": null,
    "tags": [
      "ECG",
      "12-lead",
      "EHR-linked"
    ],
    "category": "signals"
  },
  {
    "id": "chapman-shaoxing",
    "name": "Chapman–Shaoxing 12-lead ECG Database",
    "description": "12-lead ECGs from >45,000 patients at Shaoxing People's Hospital with rhythm and diagnostic labels from expert physicians.",
    "provider": "Chapman University / Shaoxing People's Hospital",
    "region": "China",
    "access": "open",
    "access_note": "Direct download from PhysioNet.",
    "url": "https://physionet.org/content/ecg-arrhythmia/",
    "paper": {
      "label": "Zheng J et al. Sci Data 2020;7:48",
      "url": "https://doi.org/10.1038/s41597-020-0386-x"
    },
    "tags": [
      "ECG",
      "12-lead",
      "Asia"
    ],
    "category": "signals"
  },
  {
    "id": "code-15",
    "name": "CODE-15%",
    "description": "345,779 12-lead ECGs from 233,770 Brazilian patients (a 15% sample of the CODE telehealth cohort) with labels for six abnormalities and mortality follow-up.",
    "provider": "Telehealth Network of Minas Gerais / Uppsala University",
    "region": "Brazil",
    "access": "open",
    "access_note": "Direct download from Zenodo (CC BY 4.0).",
    "url": "https://zenodo.org/records/4916206",
    "paper": {
      "label": "Ribeiro AH et al. Nat Commun 2020;11:1760",
      "url": "https://doi.org/10.1038/s41467-020-15432-4"
    },
    "tags": [
      "ECG",
      "12-lead",
      "LMIC"
    ],
    "category": "signals"
  },
  {
    "id": "physionet-challenge-2021",
    "name": "PhysioNet/CinC Challenge 2020–2021 (12-lead ECG)",
    "description": "~88,000 annotated ECGs pooled from multiple international databases (CPSC, INCART, PTB, Georgia, Chapman, Ningbo) with SNOMED-CT diagnostic labels.",
    "provider": "PhysioNet / Computing in Cardiology",
    "region": "International",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://physionet.org/content/challenge-2021/",
    "paper": {
      "label": "Perez Alday EA et al. Physiol Meas 2020;41:124003",
      "url": "https://doi.org/10.1088/1361-6579/abc960"
    },
    "tags": [
      "ECG",
      "challenge",
      "multi-source"
    ],
    "category": "signals"
  },
  {
    "id": "physionet-challenge-2017",
    "name": "PhysioNet/CinC Challenge 2017 (AF)",
    "description": "8,528 short single-lead ECGs from AliveCor devices labelled normal, AF, other rhythm or noisy.",
    "provider": "PhysioNet / Computing in Cardiology",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://physionet.org/content/challenge-2017/",
    "paper": {
      "label": "Clifford GD et al. Comput Cardiol 2017;44",
      "url": "https://doi.org/10.22489/CinC.2017.065-469"
    },
    "tags": [
      "ECG",
      "atrial fibrillation",
      "mobile"
    ],
    "category": "signals"
  },
  {
    "id": "icentia11k",
    "name": "Icentia11k",
    "description": "Continuous single-lead ECG from 11,000 patients (up to two weeks each) with ~2.7 billion labelled beats — a huge resource for self-supervised ECG learning.",
    "provider": "Icentia / Mila via PhysioNet",
    "region": "Canada",
    "access": "open",
    "access_note": "Direct download (non-commercial licence).",
    "url": "https://physionet.org/content/icentia11k-continuous-ecg/",
    "paper": {
      "label": "Tan S et al. arXiv 2019",
      "url": "https://arxiv.org/abs/1910.09570"
    },
    "tags": [
      "ECG",
      "Holter",
      "large-scale"
    ],
    "category": "signals"
  },
  {
    "id": "mimic-wdb",
    "name": "MIMIC-III & MIMIC-IV Waveform Databases",
    "description": "Bedside-monitor waveforms (ECG, ABP, PPG, respiration) and numeric trends recorded from ICU patients, partly matched to the clinical databases.",
    "provider": "MIT Laboratory for Computational Physiology",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download (ODbL).",
    "url": "https://physionet.org/content/mimic4wdb/",
    "paper": null,
    "tags": [
      "waveforms",
      "ICU",
      "PPG"
    ],
    "category": "signals"
  },
  {
    "id": "bidmc-ppg",
    "name": "BIDMC PPG and Respiration Dataset",
    "description": "53 eight-minute recordings of ECG, PPG and impedance pneumography with manual breath annotations, extracted from MIMIC-II.",
    "provider": "University of Oxford / PhysioNet",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://physionet.org/content/bidmc/",
    "paper": {
      "label": "Pimentel MAF et al. IEEE Trans Biomed Eng 2017;64:1914",
      "url": "https://doi.org/10.1109/TBME.2016.2613124"
    },
    "tags": [
      "PPG",
      "respiratory rate"
    ],
    "category": "signals"
  },
  {
    "id": "sleep-edf",
    "name": "Sleep-EDF Expanded",
    "description": "197 whole-night polysomnographic recordings (EEG, EOG, chin EMG) with hypnograms scored by technicians — the standard sleep-staging benchmark.",
    "provider": "PhysioNet",
    "region": "Netherlands",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://physionet.org/content/sleep-edfx/",
    "paper": {
      "label": "Kemp B et al. IEEE Trans Biomed Eng 2000;47:1185",
      "url": "https://doi.org/10.1109/10.867928"
    },
    "tags": [
      "sleep",
      "EEG",
      "polysomnography"
    ],
    "category": "signals"
  },
  {
    "id": "nsrr",
    "name": "National Sleep Research Resource (NSRR)",
    "description": "Polysomnography and clinical data from >40,000 sleep studies across major cohorts (SHHS, MESA, CFS, CHAT, MrOS, APPLES).",
    "provider": "Brigham and Women's Hospital / NHLBI",
    "region": "USA",
    "access": "application",
    "access_note": "Free; online data access request with signed Data Access and Use Agreement.",
    "url": "https://sleepdata.org/",
    "paper": {
      "label": "Zhang GQ et al. JAMIA 2018;25:1351",
      "url": "https://doi.org/10.1093/jamia/ocy064"
    },
    "tags": [
      "sleep",
      "polysomnography",
      "cohort"
    ],
    "category": "signals"
  },
  {
    "id": "apnea-ecg",
    "name": "Apnea-ECG Database",
    "description": "70 overnight single-lead ECG recordings with minute-by-minute apnoea annotations for ECG-based sleep-apnoea detection.",
    "provider": "PhysioNet",
    "region": "Germany",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://physionet.org/content/apnea-ecg/",
    "paper": null,
    "tags": [
      "sleep apnoea",
      "ECG"
    ],
    "category": "signals"
  },
  {
    "id": "ctu-uhb",
    "name": "CTU-UHB Intrapartum Cardiotocography Database",
    "description": "552 intrapartum CTG recordings (fetal heart rate and uterine contractions) with umbilical-artery pH and clinical outcomes from Brno University Hospital.",
    "provider": "Czech Technical University / University Hospital Brno",
    "region": "Czech Republic",
    "access": "open",
    "access_note": "Direct download from PhysioNet.",
    "url": "https://physionet.org/content/ctu-uhb-ctgdb/",
    "paper": {
      "label": "Chudáček V et al. BMC Pregnancy Childbirth 2014;14:16",
      "url": "https://doi.org/10.1186/1471-2393-14-16"
    },
    "tags": [
      "obstetrics",
      "CTG",
      "fetal monitoring"
    ],
    "category": "signals"
  },
  {
    "id": "physionet-challenge-2016",
    "name": "PhysioNet/CinC Challenge 2016 (Heart Sounds)",
    "description": "3,000+ phonocardiogram recordings from multiple databases labelled normal or abnormal.",
    "provider": "PhysioNet / Computing in Cardiology",
    "region": "International",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://physionet.org/content/challenge-2016/",
    "paper": {
      "label": "Liu C et al. Physiol Meas 2016;37:2181",
      "url": "https://doi.org/10.1088/0967-3334/37/12/2181"
    },
    "tags": [
      "phonocardiogram",
      "heart sounds"
    ],
    "category": "signals"
  },
  {
    "id": "circor",
    "name": "CirCor DigiScope Phonocardiogram Dataset",
    "description": "5,272 heart-sound recordings from 1,568 children in Brazil with murmur annotations (timing, grade, shape) by experts.",
    "provider": "CirCor / PhysioNet",
    "region": "Brazil",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://physionet.org/content/circor-heart-sound/",
    "paper": {
      "label": "Oliveira J et al. IEEE J Biomed Health Inform 2022;26:2524",
      "url": "https://doi.org/10.1109/JBHI.2021.3137048"
    },
    "tags": [
      "heart sounds",
      "paediatrics",
      "LMIC"
    ],
    "category": "signals"
  },
  {
    "id": "icbhi",
    "name": "ICBHI 2017 Respiratory Sound Database",
    "description": "920 annotated lung-sound recordings from 126 patients (healthy, COPD, asthma, pneumonia, etc.) with crackle and wheeze labels.",
    "provider": "ICBHI / Aristotle University of Thessaloniki & University of Coimbra",
    "region": "Greece / Portugal",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://bhichallenge.med.auth.gr/",
    "paper": {
      "label": "Rocha BM et al. Physiol Meas 2019;40:035001",
      "url": "https://doi.org/10.1088/1361-6579/ab03ea"
    },
    "tags": [
      "lung sounds",
      "respiratory"
    ],
    "category": "signals"
  },
  {
    "id": "coughvid",
    "name": "COUGHVID",
    "description": "25,000+ crowdsourced cough recordings with self-reported COVID-19 status and expert annotations on a subset.",
    "provider": "EPFL",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from Zenodo (CC BY 4.0).",
    "url": "https://zenodo.org/records/4048312",
    "paper": {
      "label": "Orlandic L et al. Sci Data 2021;8:156",
      "url": "https://doi.org/10.1038/s41597-021-00937-4"
    },
    "tags": [
      "cough",
      "audio",
      "COVID-19"
    ],
    "category": "signals"
  },
  {
    "id": "coswara",
    "name": "Coswara",
    "description": "Breathing, cough, vowel and counting audio from ~2,700 Indian participants with COVID-19 status and symptoms.",
    "provider": "Indian Institute of Science, Bangalore",
    "region": "India",
    "access": "open",
    "access_note": "Direct download from GitHub (CC BY 4.0).",
    "url": "https://github.com/iiscleap/Coswara-Data",
    "paper": {
      "label": "Bhattacharya D et al. Sci Data 2023;10:397",
      "url": "https://doi.org/10.1038/s41597-023-02266-0"
    },
    "tags": [
      "audio",
      "COVID-19",
      "LMIC"
    ],
    "category": "signals"
  },
  {
    "id": "wesad",
    "name": "WESAD",
    "description": "Wrist- and chest-worn multimodal sensor data (BVP, EDA, ECG, EMG, respiration, temperature, accelerometry) from 15 subjects during stress and amusement.",
    "provider": "Bosch / University of Siegen",
    "region": "Germany",
    "access": "open",
    "access_note": "Direct download (UCI repository).",
    "url": "https://archive.ics.uci.edu/dataset/465/wesad+wearable+stress+and+affect+detection",
    "paper": {
      "label": "Schmidt P et al. ICMI 2018",
      "url": "https://doi.org/10.1145/3242969.3242985"
    },
    "tags": [
      "wearables",
      "stress",
      "EDA"
    ],
    "category": "signals"
  },
  {
    "id": "ppg-dalia",
    "name": "PPG-DaLiA",
    "description": "Wrist PPG, accelerometer and chest ECG from 15 subjects during daily-life activities for motion-robust heart-rate estimation.",
    "provider": "Bosch / University of Siegen",
    "region": "Germany",
    "access": "open",
    "access_note": "Direct download (UCI repository).",
    "url": "https://archive.ics.uci.edu/dataset/495/ppg+dalia",
    "paper": {
      "label": "Reiss A et al. Sensors 2019;19:3079",
      "url": "https://doi.org/10.3390/s19143079"
    },
    "tags": [
      "wearables",
      "PPG",
      "heart rate"
    ],
    "category": "signals"
  },
  {
    "id": "lifesnaps",
    "name": "LifeSnaps",
    "description": "Four months of Fitbit data (steps, sleep, heart rate, SpO2, temperature) plus surveys and ecological momentary assessments from 71 participants in the wild.",
    "provider": "Aristotle University of Thessaloniki",
    "region": "Europe",
    "access": "open",
    "access_note": "Direct download from Zenodo.",
    "url": "https://zenodo.org/records/7229547",
    "paper": {
      "label": "Yfantidou S et al. Sci Data 2022;9:663",
      "url": "https://doi.org/10.1038/s41597-022-01764-x"
    },
    "tags": [
      "wearables",
      "Fitbit",
      "sleep"
    ],
    "category": "signals"
  },
  {
    "id": "uci-har",
    "name": "UCI Human Activity Recognition (Smartphones)",
    "description": "Smartphone accelerometer and gyroscope signals from 30 volunteers performing six daily activities — a classic activity-recognition dataset.",
    "provider": "UCI Machine Learning Repository",
    "region": "Italy / Spain",
    "access": "open",
    "access_note": "Direct download (CC BY 4.0).",
    "url": "https://archive.ics.uci.edu/dataset/240/human+activity+recognition+using+smartphones",
    "paper": null,
    "tags": [
      "activity recognition",
      "smartphone"
    ],
    "category": "signals"
  },
  {
    "id": "mimic-iii-ext-ppg",
    "name": "MIMIC-III-Ext-PPG",
    "description": "4.9 million 30-second PPG segments from MIMIC-III ICU patients with rhythm labels, for PPG-based arrhythmia research.",
    "provider": "PhysioNet",
    "region": "USA",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + CITI training + DUA.",
    "url": "https://physionet.org/content/mimic-iii-ext-ppg/",
    "paper": null,
    "tags": [
      "PPG",
      "ICU",
      "arrhythmia"
    ],
    "category": "signals"
  },
  {
    "id": "echonext",
    "name": "EchoNext",
    "description": "100,000 12-lead ECGs from Columbia University Irving Medical Center paired with echocardiography-derived structural heart disease labels.",
    "provider": "Columbia University / NewYork-Presbyterian",
    "region": "USA",
    "access": "registration",
    "access_note": "PhysioNet account + signed data use agreement (restricted access).",
    "url": "https://physionet.org/content/echonext/",
    "paper": {
      "label": "Poterucha TJ et al. Nature 2025",
      "url": "https://doi.org/10.1038/s41586-025-09227-0"
    },
    "tags": [
      "ECG",
      "structural heart disease"
    ],
    "category": "signals"
  },
  {
    "id": "b2ai-voice",
    "name": "Bridge2AI-Voice",
    "description": "Voice and speech recordings with clinical data from participants with voice, neurological, respiratory and mood disorders (NIH Bridge2AI).",
    "provider": "Bridge2AI-Voice Consortium",
    "region": "USA / Canada",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + DUA (derived features may be open).",
    "url": "https://physionet.org/content/b2ai-voice/",
    "paper": null,
    "tags": [
      "voice",
      "audio",
      "biomarkers"
    ],
    "category": "signals"
  },
  {
    "id": "covid-19-sounds",
    "name": "COVID-19 Sounds",
    "description": "53,449 breathing, cough and voice recordings from 36,116 participants with COVID-19 test status and symptoms.",
    "provider": "University of Cambridge",
    "region": "International",
    "access": "application",
    "access_note": "Free; sign a data transfer agreement with Cambridge.",
    "url": "https://www.covid-19-sounds.org/en/blog/neurips_dataset.html",
    "paper": {
      "label": "Xia T et al. NeurIPS 2021 Datasets & Benchmarks",
      "url": "https://openreview.net/forum?id=9KArJb4r5ZQ"
    },
    "tags": [
      "audio",
      "COVID-19"
    ],
    "category": "signals"
  },
  {
    "id": "pads-parkinsons",
    "name": "PADS (Parkinson's Disease Smartwatch)",
    "description": "Smartwatch accelerometer and gyroscope recordings from 469 participants (Parkinson's, differential diagnoses, healthy) during structured assessments.",
    "provider": "University of Münster",
    "region": "Germany",
    "access": "open",
    "access_note": "Direct download from PhysioNet.",
    "url": "https://physionet.org/content/parkinsons-disease-smartwatch/",
    "paper": null,
    "tags": [
      "wearables",
      "Parkinson's"
    ],
    "category": "signals"
  },
  {
    "id": "kaggle-parkinsons",
    "name": "Parkinson's Kaggle Challenges (AMP-PD & Freezing of Gait)",
    "description": "AMP-PD: longitudinal CSF protein/peptide data for progression prediction; FoG: lower-back accelerometry from lab, home and daily-living settings for freezing-of-gait detection.",
    "provider": "Accelerating Medicines Partnership / Michael J. Fox Foundation",
    "region": "International",
    "access": "registration",
    "access_note": "Free Kaggle account and acceptance of competition rules.",
    "url": "https://www.kaggle.com/competitions/tlvmc-parkinsons-freezing-gait-prediction",
    "paper": null,
    "tags": [
      "Parkinson's",
      "wearables",
      "proteomics"
    ],
    "category": "signals"
  },
  {
    "id": "google-rppg",
    "name": "Google Heart Rate Measurement (rPPG) Study Dataset",
    "description": "Facial videos with skin-tone measurements and reference heart rate for evaluating camera-based (remote PPG) heart-rate measurement across skin tones.",
    "provider": "Google Health",
    "region": "USA",
    "access": "open",
    "access_note": "Details and download instructions on GitHub.",
    "url": "https://github.com/Google-Health/consumer-health-research/tree/main/rppg",
    "paper": null,
    "tags": [
      "rPPG",
      "video",
      "skin tone"
    ],
    "category": "signals"
  },
  {
    "id": "propofol-anesthesia",
    "name": "Propofol Anesthesia Dynamics",
    "description": "ECG and autonomic recordings from nine volunteers during propofol-induced loss and recovery of consciousness.",
    "provider": "MIT / MGH via PhysioNet",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from PhysioNet.",
    "url": "https://physionet.org/content/propofol-anesthesia-dynamics/",
    "paper": null,
    "tags": [
      "anaesthesia",
      "ECG",
      "autonomic"
    ],
    "category": "signals"
  },
  {
    "id": "dcsm-sleep",
    "name": "DCSM Sleep Staging Dataset",
    "description": "255 anonymised overnight polysomnography recordings from the Danish Center for Sleep Medicine with expert sleep stages.",
    "provider": "University of Copenhagen",
    "region": "Denmark",
    "access": "open",
    "access_note": "Direct download from ERDA.",
    "url": "https://erda.ku.dk/public/archives/db553715ecbe1f3ac66c1dc569826eef/published-archive.html",
    "paper": null,
    "tags": [
      "sleep",
      "polysomnography"
    ],
    "category": "signals"
  },
  {
    "id": "dreem-open",
    "name": "Dreem Open Datasets (DOD-H & DOD-O)",
    "description": "Polysomnography from 25 healthy adults and 55 obstructive sleep apnoea patients, each scored by five sleep technologists.",
    "provider": "Dreem",
    "region": "France",
    "access": "open",
    "access_note": "Direct download via GitHub instructions.",
    "url": "https://github.com/Dreem-Organization/dreem-learning-open",
    "paper": {
      "label": "Guillot A et al. IEEE Trans Neural Syst Rehabil Eng 2020;28:1955",
      "url": "https://doi.org/10.1109/TNSRE.2020.3011181"
    },
    "tags": [
      "sleep",
      "polysomnography",
      "sleep apnoea"
    ],
    "category": "signals"
  },
  {
    "id": "gdc",
    "name": "NCI Genomic Data Commons (TCGA & more)",
    "description": "Harmonised genomic and clinical data from TCGA (33 cancer types, ~11,000 patients), TARGET, CPTAC and other NCI programmes, with web portal and API.",
    "provider": "US National Cancer Institute",
    "region": "USA / International",
    "access": "open",
    "access_note": "Open tier (clinical, expression, somatic mutation calls) freely downloadable; raw reads and germline data need dbGaP authorisation.",
    "url": "https://portal.gdc.cancer.gov/",
    "paper": {
      "label": "Grossman RL et al. N Engl J Med 2016;375:1109",
      "url": "https://doi.org/10.1056/NEJMp1607591"
    },
    "tags": [
      "cancer genomics",
      "TCGA",
      "API"
    ],
    "category": "omics"
  },
  {
    "id": "cbioportal",
    "name": "cBioPortal for Cancer Genomics",
    "description": "Interactive portal for exploring and downloading processed multidimensional cancer genomics datasets from 400+ studies, with clinical annotations.",
    "provider": "Memorial Sloan Kettering / Dana-Farber / Princess Margaret",
    "region": "International",
    "access": "open",
    "access_note": "Open web, downloads and API.",
    "url": "https://www.cbioportal.org/",
    "paper": {
      "label": "Cerami E et al. Cancer Discov 2012;2:401",
      "url": "https://doi.org/10.1158/2159-8290.CD-12-0095"
    },
    "tags": [
      "cancer genomics",
      "visualisation"
    ],
    "category": "omics"
  },
  {
    "id": "icgc-argo",
    "name": "ICGC ARGO",
    "description": "International Cancer Genome Consortium platform with clinically annotated tumour genomes from >100,000 patients (successor to the PCAWG/ICGC 25K data).",
    "provider": "International Cancer Genome Consortium",
    "region": "International",
    "access": "registration",
    "access_note": "Open-tier summary data browsable; controlled data require application through DACO.",
    "url": "https://platform.icgc-argo.org/",
    "paper": {
      "label": "ICGC/TCGA PCAWG Consortium. Nature 2020;578:82",
      "url": "https://doi.org/10.1038/s41586-020-1969-6"
    },
    "tags": [
      "cancer genomics",
      "international"
    ],
    "category": "omics"
  },
  {
    "id": "cosmic",
    "name": "COSMIC",
    "full_name": "Catalogue Of Somatic Mutations In Cancer",
    "description": "Expert-curated database of somatic mutations in human cancer, including Cancer Gene Census, mutational signatures and drug-resistance mutations.",
    "provider": "Wellcome Sanger Institute",
    "region": "International",
    "access": "registration",
    "access_note": "Free for academic/non-profit use with registration; commercial use licensed.",
    "url": "https://cancer.sanger.ac.uk/cosmic",
    "paper": {
      "label": "Tate JG et al. Nucleic Acids Res 2019;47:D941",
      "url": "https://doi.org/10.1093/nar/gky1015"
    },
    "tags": [
      "cancer",
      "mutations",
      "curated"
    ],
    "category": "omics"
  },
  {
    "id": "1000-genomes",
    "name": "1000 Genomes / IGSR",
    "description": "Whole-genome sequences of 3,202 individuals from 26 populations, the reference catalogue of human genetic variation maintained by the International Genome Sample Resource.",
    "provider": "EMBL-EBI",
    "region": "Global",
    "access": "open",
    "access_note": "Direct download (no restrictions).",
    "url": "https://www.internationalgenome.org/",
    "paper": {
      "label": "1000 Genomes Project Consortium. Nature 2015;526:68",
      "url": "https://doi.org/10.1038/nature15393"
    },
    "tags": [
      "population genetics",
      "WGS"
    ],
    "category": "omics"
  },
  {
    "id": "gnomad",
    "name": "gnomAD",
    "full_name": "Genome Aggregation Database",
    "description": "Allele frequencies and constraint metrics from >800,000 exomes and genomes across diverse ancestries — essential for variant interpretation.",
    "provider": "Broad Institute",
    "region": "Global",
    "access": "open",
    "access_note": "Open browser and downloads (VCF, Hail tables) on public clouds.",
    "url": "https://gnomad.broadinstitute.org/",
    "paper": {
      "label": "Karczewski KJ et al. Nature 2020;581:434",
      "url": "https://doi.org/10.1038/s41586-020-2308-7"
    },
    "tags": [
      "variant frequencies",
      "constraint"
    ],
    "category": "omics"
  },
  {
    "id": "topmed-bravo",
    "name": "TOPMed BRAVO",
    "description": "Variant browser and allele frequencies from >150,000 deeply sequenced genomes in NHLBI's TOPMed programme; individual-level data via dbGaP.",
    "provider": "University of Michigan / NHLBI",
    "region": "USA",
    "access": "registration",
    "access_note": "Browser requires a free Google login; individual data by dbGaP application.",
    "url": "https://bravo.sph.umich.edu/",
    "paper": {
      "label": "Taliun D et al. Nature 2021;590:290",
      "url": "https://doi.org/10.1038/s41586-021-03205-y"
    },
    "tags": [
      "WGS",
      "variant frequencies"
    ],
    "category": "omics"
  },
  {
    "id": "clinvar",
    "name": "ClinVar",
    "description": "Public archive of relationships between human genetic variants and phenotypes with supporting evidence and clinical-significance assertions from labs worldwide.",
    "provider": "NCBI",
    "region": "Global",
    "access": "open",
    "access_note": "Open web, FTP and API.",
    "url": "https://www.ncbi.nlm.nih.gov/clinvar/",
    "paper": {
      "label": "Landrum MJ et al. Nucleic Acids Res 2018;46:D1062",
      "url": "https://doi.org/10.1093/nar/gkx1153"
    },
    "tags": [
      "variant interpretation",
      "clinical genetics"
    ],
    "category": "omics"
  },
  {
    "id": "gwas-catalog",
    "name": "GWAS Catalog",
    "description": "Curated collection of all published genome-wide association studies with >600,000 variant–trait associations and full summary statistics for many studies.",
    "provider": "NHGRI-EBI",
    "region": "Global",
    "access": "open",
    "access_note": "Open web, downloads and API.",
    "url": "https://www.ebi.ac.uk/gwas/",
    "paper": {
      "label": "Sollis E et al. Nucleic Acids Res 2023;51:D977",
      "url": "https://doi.org/10.1093/nar/gkac1010"
    },
    "tags": [
      "GWAS",
      "summary statistics"
    ],
    "category": "omics"
  },
  {
    "id": "pgs-catalog",
    "name": "PGS Catalog",
    "description": "Open database of published polygenic scores with variant weights, metadata and performance metrics for thousands of traits.",
    "provider": "University of Cambridge / EMBL-EBI",
    "region": "Global",
    "access": "open",
    "access_note": "Open downloads and API.",
    "url": "https://www.pgscatalog.org/",
    "paper": {
      "label": "Lambert SA et al. Nat Genet 2021;53:420",
      "url": "https://doi.org/10.1038/s41588-021-00783-5"
    },
    "tags": [
      "polygenic scores"
    ],
    "category": "omics"
  },
  {
    "id": "finngen",
    "name": "FinnGen (summary statistics)",
    "description": "GWAS summary statistics for ~2,400 disease endpoints from ~500,000 Finnish biobank participants linked to national health registries.",
    "provider": "FinnGen / University of Helsinki",
    "region": "Finland",
    "access": "open",
    "access_note": "Summary statistics free to download after filling a short form; individual data not shared.",
    "url": "https://www.finngen.fi/en/access_results",
    "paper": {
      "label": "Kurki MI et al. Nature 2023;613:508",
      "url": "https://doi.org/10.1038/s41586-022-05473-8"
    },
    "tags": [
      "GWAS",
      "biobank",
      "summary statistics"
    ],
    "category": "omics"
  },
  {
    "id": "pan-ukbb",
    "name": "Pan-UK Biobank (summary statistics)",
    "description": "Multi-ancestry GWAS summary statistics for >7,000 phenotypes in UK Biobank across six continental ancestry groups — free, even though UK Biobank itself is fee-based.",
    "provider": "Broad Institute",
    "region": "United Kingdom",
    "access": "open",
    "access_note": "Open download on AWS.",
    "url": "https://pan.ukbb.broadinstitute.org/",
    "paper": null,
    "tags": [
      "GWAS",
      "summary statistics",
      "multi-ancestry"
    ],
    "category": "omics"
  },
  {
    "id": "gtex",
    "name": "GTEx Portal",
    "description": "Gene expression and eQTLs across 54 non-diseased tissues from ~1,000 post-mortem donors, showing how genetic variation regulates expression.",
    "provider": "Broad Institute / NIH Common Fund",
    "region": "USA",
    "access": "open",
    "access_note": "Expression and eQTL results open; raw genotypes and reads via dbGaP.",
    "url": "https://gtexportal.org/",
    "paper": {
      "label": "GTEx Consortium. Science 2020;369:1318",
      "url": "https://doi.org/10.1126/science.aaz1776"
    },
    "tags": [
      "transcriptomics",
      "eQTL"
    ],
    "category": "omics"
  },
  {
    "id": "encode",
    "name": "ENCODE",
    "description": "Encyclopedia of DNA Elements: functional genomics data (ChIP-seq, ATAC-seq, RNA-seq, Hi-C) mapping regulatory elements across human and mouse cell types.",
    "provider": "NHGRI ENCODE Consortium",
    "region": "International",
    "access": "open",
    "access_note": "Open (no restrictions).",
    "url": "https://www.encodeproject.org/",
    "paper": {
      "label": "ENCODE Project Consortium. Nature 2012;489:57",
      "url": "https://doi.org/10.1038/nature11247"
    },
    "tags": [
      "functional genomics",
      "epigenomics"
    ],
    "category": "omics"
  },
  {
    "id": "geo",
    "name": "Gene Expression Omnibus (GEO)",
    "description": "NCBI's public repository of >200,000 functional-genomics studies (microarray, RNA-seq, methylation, single-cell) submitted with publications.",
    "provider": "NCBI",
    "region": "Global",
    "access": "open",
    "access_note": "Open; GEO2R provides in-browser analysis.",
    "url": "https://www.ncbi.nlm.nih.gov/geo/",
    "paper": {
      "label": "Barrett T et al. Nucleic Acids Res 2013;41:D991",
      "url": "https://doi.org/10.1093/nar/gks1193"
    },
    "tags": [
      "transcriptomics",
      "repository"
    ],
    "category": "omics"
  },
  {
    "id": "arrayexpress",
    "name": "ArrayExpress (BioStudies)",
    "description": "EMBL-EBI archive of functional-genomics experiments — Europe's counterpart to GEO — now served through the BioStudies platform.",
    "provider": "EMBL-EBI",
    "region": "Global",
    "access": "open",
    "access_note": "Open.",
    "url": "https://www.ebi.ac.uk/biostudies/arrayexpress",
    "paper": null,
    "tags": [
      "transcriptomics",
      "repository"
    ],
    "category": "omics"
  },
  {
    "id": "sra",
    "name": "Sequence Read Archive (SRA)",
    "description": "The largest public repository of raw high-throughput sequencing reads, with cloud-hosted copies on AWS and Google Cloud.",
    "provider": "NCBI",
    "region": "Global",
    "access": "open",
    "access_note": "Open for most data; human controlled-access reads via dbGaP.",
    "url": "https://www.ncbi.nlm.nih.gov/sra",
    "paper": null,
    "tags": [
      "sequencing",
      "raw data"
    ],
    "category": "omics"
  },
  {
    "id": "dbgap",
    "name": "dbGaP",
    "full_name": "database of Genotypes and Phenotypes",
    "description": "NIH archive of individual-level genotype, sequence and phenotype data from >1,500 studies (e.g. Framingham, TOPMed, GTEx raw data).",
    "provider": "NCBI",
    "region": "USA",
    "access": "application",
    "access_note": "Free; eRA Commons login and Data Access Committee approval per study. Public summaries open.",
    "url": "https://www.ncbi.nlm.nih.gov/gap/",
    "paper": {
      "label": "Mailman MD et al. Nat Genet 2007;39:1181",
      "url": "https://doi.org/10.1038/ng1007-1181"
    },
    "tags": [
      "genotypes",
      "controlled access"
    ],
    "category": "omics"
  },
  {
    "id": "ega",
    "name": "European Genome-phenome Archive (EGA)",
    "description": "Repository of controlled-access human genetic and phenotypic data from biomedical studies worldwide (e.g. H3Africa, many cancer cohorts).",
    "provider": "EMBL-EBI / CRG",
    "region": "Global",
    "access": "application",
    "access_note": "Free; request access from each dataset's Data Access Committee.",
    "url": "https://ega-archive.org/",
    "paper": {
      "label": "Freeberg MA et al. Nucleic Acids Res 2022;50:D980",
      "url": "https://doi.org/10.1093/nar/gkab1059"
    },
    "tags": [
      "genomics",
      "controlled access"
    ],
    "category": "omics"
  },
  {
    "id": "hca",
    "name": "Human Cell Atlas Data Portal",
    "description": "Single-cell and spatial transcriptomics data from millions of cells across human tissues, contributing to a reference map of every cell type.",
    "provider": "Human Cell Atlas consortium",
    "region": "Global",
    "access": "open",
    "access_note": "Open (managed-access projects flagged separately).",
    "url": "https://data.humancellatlas.org/",
    "paper": {
      "label": "Regev A et al. eLife 2017;6:e27041",
      "url": "https://doi.org/10.7554/eLife.27041"
    },
    "tags": [
      "single-cell",
      "atlas"
    ],
    "category": "omics"
  },
  {
    "id": "cellxgene",
    "name": "CZ CELLxGENE Discover",
    "description": "Curated, standardised collection of 1,000+ single-cell datasets (100M+ cells) with in-browser exploration and a Python/R Census API.",
    "provider": "Chan Zuckerberg Initiative",
    "region": "Global",
    "access": "open",
    "access_note": "Open (mostly CC BY 4.0).",
    "url": "https://cellxgene.cziscience.com/",
    "paper": null,
    "tags": [
      "single-cell",
      "API"
    ],
    "category": "omics"
  },
  {
    "id": "hpa",
    "name": "Human Protein Atlas",
    "description": "Maps of protein expression across human tissues, cells, organelles, blood and cancers using antibody-based imaging and transcriptomics.",
    "provider": "KTH / SciLifeLab, Sweden",
    "region": "Global",
    "access": "open",
    "access_note": "Open (CC BY-SA 3.0).",
    "url": "https://www.proteinatlas.org/",
    "paper": {
      "label": "Uhlén M et al. Science 2015;347:1260419",
      "url": "https://doi.org/10.1126/science.1260419"
    },
    "tags": [
      "proteomics",
      "tissue atlas"
    ],
    "category": "omics"
  },
  {
    "id": "pdc",
    "name": "Proteomic Data Commons (CPTAC)",
    "description": "Mass-spectrometry proteomics and phosphoproteomics data from CPTAC tumour cohorts, harmonised and linked to GDC genomic data.",
    "provider": "US National Cancer Institute",
    "region": "USA",
    "access": "open",
    "access_note": "Open.",
    "url": "https://pdc.cancer.gov/",
    "paper": null,
    "tags": [
      "proteomics",
      "cancer"
    ],
    "category": "omics"
  },
  {
    "id": "pride",
    "name": "PRIDE Archive",
    "description": "The world's largest repository of mass-spectrometry proteomics data, part of the ProteomeXchange consortium.",
    "provider": "EMBL-EBI",
    "region": "Global",
    "access": "open",
    "access_note": "Open.",
    "url": "https://www.ebi.ac.uk/pride/",
    "paper": {
      "label": "Perez-Riverol Y et al. Nucleic Acids Res 2022;50:D543",
      "url": "https://doi.org/10.1093/nar/gkab1038"
    },
    "tags": [
      "proteomics",
      "repository"
    ],
    "category": "omics"
  },
  {
    "id": "metabolights",
    "name": "MetaboLights",
    "description": "Cross-species, cross-technique repository for metabolomics experiments with raw spectra, metadata and reference metabolite information.",
    "provider": "EMBL-EBI",
    "region": "Global",
    "access": "open",
    "access_note": "Open.",
    "url": "https://www.ebi.ac.uk/metabolights/",
    "paper": {
      "label": "Yurekten O et al. Nucleic Acids Res 2024;52:D640",
      "url": "https://doi.org/10.1093/nar/gkad1045"
    },
    "tags": [
      "metabolomics",
      "repository"
    ],
    "category": "omics"
  },
  {
    "id": "hmp",
    "name": "Human Microbiome Project",
    "description": "Metagenomic and 16S data from body sites of healthy adults (HMP1) and multi-omic disease cohorts (iHMP: IBD, prediabetes, pregnancy).",
    "provider": "NIH Common Fund",
    "region": "USA",
    "access": "open",
    "access_note": "Open (human reads via dbGaP).",
    "url": "https://www.hmpdacc.org/",
    "paper": {
      "label": "Human Microbiome Project Consortium. Nature 2012;486:207",
      "url": "https://doi.org/10.1038/nature11234"
    },
    "tags": [
      "microbiome",
      "metagenomics"
    ],
    "category": "omics"
  },
  {
    "id": "depmap",
    "name": "DepMap Portal",
    "description": "Cancer Dependency Map: CRISPR/RNAi gene-dependency screens, drug sensitivity and multi-omic profiles for 1,000+ cancer cell lines (incl. CCLE).",
    "provider": "Broad Institute",
    "region": "Global",
    "access": "open",
    "access_note": "Open (CC BY 4.0).",
    "url": "https://depmap.org/portal/",
    "paper": {
      "label": "Tsherniak A et al. Cell 2017;170:564",
      "url": "https://doi.org/10.1016/j.cell.2017.06.010"
    },
    "tags": [
      "cell lines",
      "CRISPR",
      "drug response"
    ],
    "category": "omics"
  },
  {
    "id": "gdsc",
    "name": "GDSC",
    "full_name": "Genomics of Drug Sensitivity in Cancer",
    "description": "Drug-response data for ~1,000 cancer cell lines against hundreds of anticancer compounds, with genomic markers of sensitivity.",
    "provider": "Wellcome Sanger Institute",
    "region": "Global",
    "access": "open",
    "access_note": "Open.",
    "url": "https://www.cancerrxgene.org/",
    "paper": {
      "label": "Yang W et al. Nucleic Acids Res 2013;41:D955",
      "url": "https://doi.org/10.1093/nar/gks1111"
    },
    "tags": [
      "pharmacogenomics",
      "cell lines"
    ],
    "category": "omics"
  },
  {
    "id": "open-targets",
    "name": "Open Targets Platform",
    "description": "Integrates genetics, omics, literature and drug data to score and prioritise target–disease associations for drug discovery.",
    "provider": "Open Targets (EMBL-EBI, Wellcome Sanger, industry partners)",
    "region": "Global",
    "access": "open",
    "access_note": "Open web, GraphQL API and bulk downloads.",
    "url": "https://platform.opentargets.org/",
    "paper": {
      "label": "Ochoa D et al. Nucleic Acids Res 2023;51:D1353",
      "url": "https://doi.org/10.1093/nar/gkac1046"
    },
    "tags": [
      "drug targets",
      "integration"
    ],
    "category": "omics"
  },
  {
    "id": "immport",
    "name": "ImmPort",
    "description": "Repository of immunology study data (clinical trials and mechanistic studies) from NIAID-funded research, curated into a standard model.",
    "provider": "NIAID",
    "region": "USA",
    "access": "registration",
    "access_note": "Free registration and acceptance of the data-use agreement.",
    "url": "https://www.immport.org/",
    "paper": {
      "label": "Bhattacharya S et al. Sci Data 2018;5:180015",
      "url": "https://doi.org/10.1038/sdata.2018.15"
    },
    "tags": [
      "immunology",
      "trials"
    ],
    "category": "omics"
  },
  {
    "id": "decipher",
    "name": "DECIPHER",
    "description": "Database of genomic variants and phenotypes from ~50,000 consented patients with rare developmental disorders, used for variant interpretation.",
    "provider": "Wellcome Sanger Institute",
    "region": "Global",
    "access": "registration",
    "access_note": "Open-access records browsable; bulk download under a data access agreement.",
    "url": "https://www.deciphergenomics.org/",
    "paper": {
      "label": "Firth HV et al. Am J Hum Genet 2009;84:524",
      "url": "https://doi.org/10.1016/j.ajhg.2009.03.010"
    },
    "tags": [
      "rare disease",
      "CNV",
      "variant interpretation"
    ],
    "category": "omics"
  },
  {
    "id": "pgc-gwas",
    "name": "PGC Psychiatric GWAS Summary Statistics",
    "description": "Genome-wide association summary statistics for schizophrenia, bipolar disorder, depression, ADHD, autism and other psychiatric disorders.",
    "provider": "Psychiatric Genomics Consortium",
    "region": "International",
    "access": "open",
    "access_note": "Direct download (most results); also mirrored on Hugging Face.",
    "url": "https://pgc.unc.edu/for-researchers/download-results/",
    "paper": null,
    "tags": [
      "GWAS",
      "psychiatry",
      "summary statistics"
    ],
    "category": "omics"
  },
  {
    "id": "tahoe-100m",
    "name": "Tahoe-100M",
    "description": "100 million single-cell transcriptomic profiles of cancer cell lines under ~1,100 drug perturbations — the largest single-cell perturbation atlas.",
    "provider": "Tahoe Therapeutics / Arc Institute",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from Hugging Face.",
    "url": "https://huggingface.co/datasets/tahoebio/Tahoe-100M",
    "paper": null,
    "tags": [
      "single-cell",
      "perturbation",
      "drug response"
    ],
    "category": "omics"
  },
  {
    "id": "clinicaltrials-gov",
    "name": "ClinicalTrials.gov",
    "description": "Registry and results database of >500,000 clinical studies in 220+ countries, with a modern REST API and bulk downloads.",
    "provider": "US National Library of Medicine",
    "region": "Global",
    "access": "open",
    "access_note": "Open web, API and bulk download.",
    "url": "https://clinicaltrials.gov/",
    "paper": {
      "label": "Zarin DA et al. N Engl J Med 2011;364:852",
      "url": "https://doi.org/10.1056/NEJMsa1012065"
    },
    "tags": [
      "registry",
      "API"
    ],
    "category": "trials"
  },
  {
    "id": "aact",
    "name": "AACT",
    "full_name": "Aggregate Analysis of ClinicalTrials.gov",
    "description": "A relational (PostgreSQL) copy of the entire ClinicalTrials.gov database refreshed daily, ideal for research on trial design and reporting.",
    "provider": "Clinical Trials Transformation Initiative (Duke)",
    "region": "Global",
    "access": "open",
    "access_note": "Flat files and database dumps open; free account needed for the live cloud database.",
    "url": "https://aact.ctti-clinicaltrials.org/",
    "paper": {
      "label": "Tasneem A et al. PLoS One 2012;7:e33677",
      "url": "https://doi.org/10.1371/journal.pone.0033677"
    },
    "tags": [
      "registry",
      "SQL",
      "meta-research"
    ],
    "category": "trials"
  },
  {
    "id": "who-ictrp",
    "name": "WHO ICTRP",
    "full_name": "International Clinical Trials Registry Platform",
    "description": "Single search portal over primary trial registries worldwide (incl. CTRI, ChiCTR, EU CTR, ISRCTN, ANZCTR, PACTR).",
    "provider": "World Health Organization",
    "region": "Global",
    "access": "open",
    "access_note": "Open search; full data export on request.",
    "url": "https://trialsearch.who.int/",
    "paper": null,
    "tags": [
      "registry",
      "global"
    ],
    "category": "trials"
  },
  {
    "id": "vivli",
    "name": "Vivli",
    "description": "Independent non-profit platform to search and request de-identified individual participant data from 7,000+ trials contributed by pharma, academia and funders (now also hosting GSK's former CSDR requests).",
    "provider": "Vivli Inc.",
    "region": "Global",
    "access": "application",
    "access_note": "Research proposal reviewed by the data contributor; analysis in a secure research environment. Generally free for academic requestors.",
    "url": "https://vivli.org/",
    "paper": {
      "label": "Bierer BE et al. N Engl J Med 2016;374:2411",
      "url": "https://doi.org/10.1056/NEJMp1605348"
    },
    "tags": [
      "IPD",
      "pharma trials"
    ],
    "category": "trials"
  },
  {
    "id": "yoda",
    "name": "YODA Project",
    "full_name": "Yale Open Data Access Project",
    "description": "Independent review and sharing of individual participant data from trials contributed by Johnson & Johnson, Medtronic, Procter & Gamble and others.",
    "provider": "Yale University",
    "region": "Global",
    "access": "application",
    "access_note": "Free; research proposal reviewed by YODA; analysis in a secure platform.",
    "url": "https://yoda.yale.edu/",
    "paper": null,
    "tags": [
      "IPD",
      "pharma trials"
    ],
    "category": "trials"
  },
  {
    "id": "project-data-sphere",
    "name": "Project Data Sphere",
    "description": "Open platform sharing patient-level data from oncology trial comparator (and some experimental) arms, with built-in analytics tools.",
    "provider": "Project Data Sphere / CEO Roundtable on Cancer",
    "region": "Global",
    "access": "registration",
    "access_note": "Free registration and acceptance of the data-sharing agreement.",
    "url": "https://www.projectdatasphere.org/",
    "paper": null,
    "tags": [
      "IPD",
      "oncology"
    ],
    "category": "trials"
  },
  {
    "id": "nctn-archive",
    "name": "NCTN/NCORP Data Archive",
    "description": "Individual-level data from completed NCI-funded cancer clinical trials run by US cooperative groups.",
    "provider": "US National Cancer Institute",
    "region": "USA",
    "access": "application",
    "access_note": "Free; data request with signed data use agreement.",
    "url": "https://nctn-data-archive.nci.nih.gov/",
    "paper": null,
    "tags": [
      "IPD",
      "oncology"
    ],
    "category": "trials"
  },
  {
    "id": "nida-datashare",
    "name": "NIDA Data Share",
    "description": "De-identified data from completed trials of the NIDA Clinical Trials Network on substance-use disorder treatments.",
    "provider": "US National Institute on Drug Abuse",
    "region": "USA",
    "access": "open",
    "access_note": "Free download after accepting the data-sharing agreement online.",
    "url": "https://datashare.nida.nih.gov/",
    "paper": null,
    "tags": [
      "IPD",
      "addiction"
    ],
    "category": "trials"
  },
  {
    "id": "n2c2",
    "name": "n2c2 / i2b2 NLP Datasets",
    "description": "Annotated de-identified clinical notes from the i2b2/n2c2 shared tasks: de-identification, medication extraction, temporal relations, obesity, smoking status, emrQA and more.",
    "provider": "Harvard DBMI",
    "region": "USA",
    "access": "application",
    "access_note": "Free; register on the DBMI Data Portal and sign the DUA.",
    "url": "https://portal.dbmi.hms.harvard.edu/projects/n2c2-nlp/",
    "paper": null,
    "tags": [
      "clinical notes",
      "NER",
      "shared tasks"
    ],
    "category": "nlp"
  },
  {
    "id": "mednli",
    "name": "MedNLI",
    "description": "14,049 clinician-annotated natural language inference pairs built from MIMIC-III notes.",
    "provider": "MIT LCP via PhysioNet",
    "region": "USA",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + DUA.",
    "url": "https://physionet.org/content/mednli/",
    "paper": {
      "label": "Romanov A, Shivade C. EMNLP 2018",
      "url": "https://doi.org/10.18653/v1/D18-1187"
    },
    "tags": [
      "NLI",
      "clinical notes"
    ],
    "category": "nlp"
  },
  {
    "id": "pubmed",
    "name": "PubMed Baseline",
    "description": "Annual XML baseline plus daily updates of 37 million+ biomedical citations and abstracts — the raw material for literature mining and biomedical language models.",
    "provider": "US National Library of Medicine",
    "region": "Global",
    "access": "open",
    "access_note": "Open FTP download; E-utilities API.",
    "url": "https://pubmed.ncbi.nlm.nih.gov/download/",
    "paper": null,
    "tags": [
      "literature",
      "abstracts"
    ],
    "category": "nlp"
  },
  {
    "id": "pmc-oa",
    "name": "PMC Open Access Subset",
    "description": "Millions of full-text journal articles from PubMed Central licensed for reuse and text mining, in XML and plain text.",
    "provider": "US National Library of Medicine",
    "region": "Global",
    "access": "open",
    "access_note": "Open bulk download (licence varies per article: commercial vs non-commercial subsets).",
    "url": "https://www.ncbi.nlm.nih.gov/pmc/tools/openftlist/",
    "paper": null,
    "tags": [
      "literature",
      "full text"
    ],
    "category": "nlp"
  },
  {
    "id": "cord-19",
    "name": "CORD-19",
    "description": "COVID-19 Open Research Dataset: ~1 million scholarly articles on COVID-19 and coronaviruses with full text, released by AI2 (final release 2022).",
    "provider": "Allen Institute for AI",
    "region": "Global",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://github.com/allenai/cord19",
    "paper": {
      "label": "Wang LL et al. ACL NLP-COVID Workshop 2020",
      "url": "https://arxiv.org/abs/2004.10706"
    },
    "tags": [
      "literature",
      "COVID-19"
    ],
    "category": "nlp"
  },
  {
    "id": "medqa",
    "name": "MedQA (USMLE)",
    "description": "Multiple-choice questions from US, Chinese and Taiwanese medical licensing exams — the most common benchmark for medical LLMs.",
    "provider": "Jin et al.",
    "region": "USA / China",
    "access": "open",
    "access_note": "Direct download (GitHub / Google Drive).",
    "url": "https://github.com/jind11/MedQA",
    "paper": {
      "label": "Jin D et al. Appl Sci 2021;11:6421",
      "url": "https://doi.org/10.3390/app11146421"
    },
    "tags": [
      "QA",
      "LLM benchmark"
    ],
    "category": "nlp"
  },
  {
    "id": "medmcqa",
    "name": "MedMCQA",
    "description": "194,000 multiple-choice questions from Indian AIIMS and NEET-PG entrance exams spanning 21 medical subjects with explanations.",
    "provider": "Saama AI Research",
    "region": "India",
    "access": "open",
    "access_note": "Direct download (MIT licence).",
    "url": "https://medmcqa.github.io/",
    "paper": {
      "label": "Pal A et al. CHIL 2022 (PMLR 174)",
      "url": "https://arxiv.org/abs/2203.14371"
    },
    "tags": [
      "QA",
      "LLM benchmark",
      "South Asia"
    ],
    "category": "nlp"
  },
  {
    "id": "pubmedqa",
    "name": "PubMedQA",
    "description": "Biomedical research question-answering dataset (yes/no/maybe) built from PubMed abstracts, with 1,000 expert-labelled instances.",
    "provider": "University of Pittsburgh / Carnegie Mellon",
    "region": "Global",
    "access": "open",
    "access_note": "Direct download (MIT licence).",
    "url": "https://pubmedqa.github.io/",
    "paper": {
      "label": "Jin Q et al. EMNLP-IJCNLP 2019",
      "url": "https://doi.org/10.18653/v1/D19-1259"
    },
    "tags": [
      "QA",
      "literature"
    ],
    "category": "nlp"
  },
  {
    "id": "bioasq",
    "name": "BioASQ",
    "description": "Annual challenge datasets for large-scale semantic indexing (MeSH) and biomedical question answering with expert-curated gold answers.",
    "provider": "BioASQ consortium",
    "region": "Global",
    "access": "registration",
    "access_note": "Free registration on the participants area.",
    "url": "http://participants-area.bioasq.org/",
    "paper": null,
    "tags": [
      "QA",
      "semantic indexing"
    ],
    "category": "nlp"
  },
  {
    "id": "mtsamples",
    "name": "MTSamples",
    "description": "~5,000 sample transcribed medical reports across 40 specialties — handy for teaching clinical NLP without any data-use agreement.",
    "provider": "MTSamples.com",
    "region": "USA",
    "access": "open",
    "access_note": "Open web access (scraped versions on Kaggle).",
    "url": "https://www.mtsamples.com/",
    "paper": null,
    "tags": [
      "clinical notes",
      "teaching"
    ],
    "category": "nlp"
  },
  {
    "id": "umls",
    "name": "UMLS Metathesaurus",
    "description": "Integrates 200+ biomedical vocabularies (SNOMED CT, ICD, MeSH, RxNorm, LOINC…) into concepts with cross-mappings, plus the Semantic Network and SPECIALIST lexicon.",
    "provider": "US National Library of Medicine",
    "region": "Global",
    "access": "registration",
    "access_note": "Free UMLS Terminology Services licence (account approval required; some source vocabularies have extra terms).",
    "url": "https://www.nlm.nih.gov/research/umls/",
    "paper": {
      "label": "Bodenreider O. Nucleic Acids Res 2004;32:D267",
      "url": "https://doi.org/10.1093/nar/gkh061"
    },
    "tags": [
      "terminology",
      "ontology"
    ],
    "category": "nlp"
  },
  {
    "id": "icd-11",
    "name": "ICD-11",
    "description": "WHO's current International Classification of Diseases with browser, coding tool and a free API for integration in health information systems.",
    "provider": "World Health Organization",
    "region": "Global",
    "access": "open",
    "access_note": "Open browser; free API registration.",
    "url": "https://icd.who.int/",
    "paper": null,
    "tags": [
      "classification",
      "API"
    ],
    "category": "nlp"
  },
  {
    "id": "loinc",
    "name": "LOINC",
    "description": "Universal code system for laboratory tests, clinical measurements and documents, required by many interoperability standards.",
    "provider": "Regenstrief Institute",
    "region": "Global",
    "access": "registration",
    "access_note": "Free account to download.",
    "url": "https://loinc.org/downloads/",
    "paper": null,
    "tags": [
      "terminology",
      "lab tests"
    ],
    "category": "nlp"
  },
  {
    "id": "mesh",
    "name": "MeSH",
    "full_name": "Medical Subject Headings",
    "description": "NLM's controlled vocabulary thesaurus used to index PubMed, available as XML, RDF and via API.",
    "provider": "US National Library of Medicine",
    "region": "Global",
    "access": "open",
    "access_note": "Open download.",
    "url": "https://www.nlm.nih.gov/databases/download/mesh.html",
    "paper": null,
    "tags": [
      "terminology",
      "indexing"
    ],
    "category": "nlp"
  },
  {
    "id": "openfda",
    "name": "openFDA",
    "description": "Public APIs and bulk downloads for FDA data: adverse event reports (drugs, devices, food), recalls, drug labels and NDC directory.",
    "provider": "US Food and Drug Administration",
    "region": "USA",
    "access": "open",
    "access_note": "Open API (free key for higher rate limits) and bulk files.",
    "url": "https://open.fda.gov/",
    "paper": {
      "label": "Kass-Hout TA et al. JAMIA 2016;23:596",
      "url": "https://doi.org/10.1093/jamia/ocv153"
    },
    "tags": [
      "pharmacovigilance",
      "API"
    ],
    "category": "drugs"
  },
  {
    "id": "faers",
    "name": "FAERS",
    "full_name": "FDA Adverse Event Reporting System",
    "description": "Quarterly raw files of spontaneous adverse-event and medication-error reports submitted to FDA since 2004 — the main source for disproportionality analyses.",
    "provider": "US Food and Drug Administration",
    "region": "USA / Global",
    "access": "open",
    "access_note": "Direct download (quarterly ASCII/XML) and public dashboard.",
    "url": "https://fis.fda.gov/extensions/FPD-QDE-FAERS/FPD-QDE-FAERS.html",
    "paper": null,
    "tags": [
      "pharmacovigilance",
      "adverse events"
    ],
    "category": "drugs"
  },
  {
    "id": "vaers",
    "name": "VAERS",
    "full_name": "Vaccine Adverse Event Reporting System",
    "description": "US national early-warning system collecting reports of adverse events after vaccination, with downloadable raw data since 1990.",
    "provider": "CDC / FDA",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download and CDC WONDER queries.",
    "url": "https://vaers.hhs.gov/data.html",
    "paper": null,
    "tags": [
      "vaccines",
      "pharmacovigilance"
    ],
    "category": "drugs"
  },
  {
    "id": "adrreports",
    "name": "EudraVigilance (adrreports.eu)",
    "description": "Public access to aggregated suspected adverse-reaction reports for medicines authorised in the European Economic Area.",
    "provider": "European Medicines Agency",
    "region": "Europe",
    "access": "open",
    "access_note": "Open web reports; line listings on request.",
    "url": "https://www.adrreports.eu/",
    "paper": null,
    "tags": [
      "pharmacovigilance",
      "Europe"
    ],
    "category": "drugs"
  },
  {
    "id": "vigiaccess",
    "name": "VigiAccess",
    "description": "Public window into VigiBase, the WHO global database of 35 million+ individual case safety reports from 150+ countries.",
    "provider": "Uppsala Monitoring Centre / WHO",
    "region": "Global",
    "access": "open",
    "access_note": "Open summary search; full VigiBase extracts require application and may carry fees.",
    "url": "https://www.vigiaccess.org/",
    "paper": null,
    "tags": [
      "pharmacovigilance",
      "global"
    ],
    "category": "drugs"
  },
  {
    "id": "nsides",
    "name": "nSIDES (OFFSIDES & TWOSIDES)",
    "description": "Data-mined drug side-effect (OFFSIDES) and drug–drug interaction (TWOSIDES) signals derived from FAERS with confounding correction.",
    "provider": "Tatonetti Lab, Columbia / Cedars-Sinai",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://nsides.io/",
    "paper": {
      "label": "Tatonetti NP et al. Sci Transl Med 2012;4:125ra31",
      "url": "https://doi.org/10.1126/scitranslmed.3003377"
    },
    "tags": [
      "drug interactions",
      "side effects"
    ],
    "category": "drugs"
  },
  {
    "id": "sider",
    "name": "SIDER",
    "description": "Side effects and indications of marketed medicines extracted from drug package inserts and mapped to MedDRA.",
    "provider": "EMBL",
    "region": "Global",
    "access": "open",
    "access_note": "Direct download (CC BY-NC-SA).",
    "url": "http://sideeffects.embl.de/",
    "paper": {
      "label": "Kuhn M et al. Nucleic Acids Res 2016;44:D1075",
      "url": "https://doi.org/10.1093/nar/gkv1075"
    },
    "tags": [
      "side effects",
      "drug labels"
    ],
    "category": "drugs"
  },
  {
    "id": "drugbank",
    "name": "DrugBank Online",
    "description": "Comprehensive drug knowledge base: chemistry, targets, pharmacology, interactions and pathways for ~16,000 drugs.",
    "provider": "OMx / University of Alberta",
    "region": "Global",
    "access": "registration",
    "access_note": "Web browsing open; full XML download free for academic use with an approved account. Commercial use licensed.",
    "url": "https://go.drugbank.com/",
    "paper": {
      "label": "Knox C et al. Nucleic Acids Res 2024;52:D1265",
      "url": "https://doi.org/10.1093/nar/gkad976"
    },
    "tags": [
      "drug knowledge base",
      "interactions"
    ],
    "category": "drugs"
  },
  {
    "id": "chembl",
    "name": "ChEMBL",
    "description": "Manually curated database of bioactive drug-like molecules with >20 million bioactivity measurements against targets.",
    "provider": "EMBL-EBI",
    "region": "Global",
    "access": "open",
    "access_note": "Open (CC BY-SA 3.0); downloads and API.",
    "url": "https://www.ebi.ac.uk/chembl/",
    "paper": {
      "label": "Zdrazil B et al. Nucleic Acids Res 2024;52:D1180",
      "url": "https://doi.org/10.1093/nar/gkad1004"
    },
    "tags": [
      "bioactivity",
      "chemistry"
    ],
    "category": "drugs"
  },
  {
    "id": "pubchem",
    "name": "PubChem",
    "description": "The world's largest free chemistry database: compounds, substances, bioassays, patents and safety information.",
    "provider": "NCBI",
    "region": "Global",
    "access": "open",
    "access_note": "Open web, FTP and PUG-REST API.",
    "url": "https://pubchem.ncbi.nlm.nih.gov/",
    "paper": {
      "label": "Kim S et al. Nucleic Acids Res 2023;51:D1373",
      "url": "https://doi.org/10.1093/nar/gkac956"
    },
    "tags": [
      "chemistry",
      "bioassays"
    ],
    "category": "drugs"
  },
  {
    "id": "dailymed",
    "name": "DailyMed",
    "description": "Official FDA-approved structured product labels (package inserts) for prescription and OTC drugs, downloadable in SPL XML.",
    "provider": "US National Library of Medicine",
    "region": "USA",
    "access": "open",
    "access_note": "Open web, API and bulk download.",
    "url": "https://dailymed.nlm.nih.gov/",
    "paper": null,
    "tags": [
      "drug labels"
    ],
    "category": "drugs"
  },
  {
    "id": "rxnorm",
    "name": "RxNorm",
    "description": "Normalised names and identifiers for clinical drugs linking many drug vocabularies, with the free RxNav APIs.",
    "provider": "US National Library of Medicine",
    "region": "USA",
    "access": "registration",
    "access_note": "Full files need a free UMLS licence; RxNav APIs are open.",
    "url": "https://www.nlm.nih.gov/research/umls/rxnorm/",
    "paper": {
      "label": "Nelson SJ et al. JAMIA 2011;18:441",
      "url": "https://doi.org/10.1136/amiajnl-2011-000116"
    },
    "tags": [
      "terminology",
      "drugs"
    ],
    "category": "drugs"
  },
  {
    "id": "atc-ddd",
    "name": "WHO ATC/DDD Index",
    "description": "The Anatomical Therapeutic Chemical classification and Defined Daily Doses used worldwide for drug-utilisation research.",
    "provider": "WHO Collaborating Centre for Drug Statistics Methodology",
    "region": "Global",
    "access": "open",
    "access_note": "Open web index (machine-readable files sold separately).",
    "url": "https://atcddd.fhi.no/atc_ddd_index/",
    "paper": null,
    "tags": [
      "classification",
      "drug utilisation"
    ],
    "category": "drugs"
  },
  {
    "id": "tdc",
    "name": "Therapeutics Data Commons (TDC)",
    "description": "66 AI-ready datasets and benchmarks spanning target discovery, ADMET, drug–drug interactions, clinical trial outcomes and more.",
    "provider": "Harvard / MIT / Stanford",
    "region": "International",
    "access": "open",
    "access_note": "Open (Python package and downloads).",
    "url": "https://tdcommons.ai/",
    "paper": {
      "label": "Huang K et al. NeurIPS 2021 Datasets & Benchmarks",
      "url": "https://arxiv.org/abs/2102.09548"
    },
    "tags": [
      "drug discovery",
      "benchmark"
    ],
    "category": "drugs"
  },
  {
    "id": "dilirank",
    "name": "DILIrank",
    "description": "1,036 FDA-approved drugs ranked by risk of drug-induced liver injury from labelling and literature.",
    "provider": "US FDA NCTR",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://www.fda.gov/science-research/liver-toxicity-knowledge-base-ltkb/drug-induced-liver-injury-rank-dilirank-dataset",
    "paper": {
      "label": "Chen M et al. Drug Discov Today 2016;21:648",
      "url": "https://doi.org/10.1016/j.drudis.2016.02.015"
    },
    "tags": [
      "hepatotoxicity",
      "drug safety"
    ],
    "category": "drugs"
  },
  {
    "id": "unitox",
    "name": "UniTox",
    "description": "LLM-extracted, clinician-checked toxicity summaries and ratings across eight organ systems for 2,418 FDA-approved drugs.",
    "provider": "Stanford University",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from Zenodo.",
    "url": "https://zenodo.org/records/11627822",
    "paper": null,
    "tags": [
      "toxicity",
      "drug labels"
    ],
    "category": "drugs"
  },
  {
    "id": "who-tb",
    "name": "WHO Global Tuberculosis Data",
    "description": "Country-reported TB notifications, treatment outcomes, drug resistance, TB/HIV, financing and WHO incidence estimates since 2000.",
    "provider": "World Health Organization",
    "region": "Global",
    "access": "open",
    "access_note": "Direct CSV download.",
    "url": "https://www.who.int/teams/global-programme-on-tuberculosis-and-lung-health/data",
    "paper": null,
    "tags": [
      "tuberculosis",
      "surveillance"
    ],
    "category": "infectious"
  },
  {
    "id": "flunet",
    "name": "WHO FluNet",
    "description": "Weekly virological surveillance data on influenza (and SARS-CoV-2/RSV) from national influenza centres in 150+ countries since 1995.",
    "provider": "World Health Organization",
    "region": "Global",
    "access": "open",
    "access_note": "Open download.",
    "url": "https://www.who.int/tools/flunet",
    "paper": null,
    "tags": [
      "influenza",
      "surveillance"
    ],
    "category": "infectious"
  },
  {
    "id": "map",
    "name": "Malaria Atlas Project",
    "description": "Geospatial estimates of malaria prevalence, incidence and intervention coverage plus parasite-rate and vector-occurrence surveys, with an R package.",
    "provider": "Telethon Kids Institute / Curtin University",
    "region": "Global",
    "access": "open",
    "access_note": "Open download (CC BY).",
    "url": "https://malariaatlas.org/",
    "paper": null,
    "tags": [
      "malaria",
      "geospatial"
    ],
    "category": "infectious"
  },
  {
    "id": "glass",
    "name": "WHO GLASS",
    "full_name": "Global Antimicrobial Resistance and Use Surveillance System",
    "description": "Country-reported surveillance data on resistance in priority bacterial pathogens and antimicrobial consumption, with an interactive dashboard.",
    "provider": "World Health Organization",
    "region": "Global",
    "access": "open",
    "access_note": "Open dashboard and downloads.",
    "url": "https://www.who.int/initiatives/glass",
    "paper": null,
    "tags": [
      "AMR",
      "antibiotic use"
    ],
    "category": "infectious"
  },
  {
    "id": "resistancemap",
    "name": "ResistanceMap",
    "description": "Interactive maps and trends of antibiotic resistance and antibiotic consumption for countries worldwide, including the Drug Resistance Index.",
    "provider": "One Health Trust (formerly CDDEP)",
    "region": "Global",
    "access": "open",
    "access_note": "Open web tool.",
    "url": "https://resistancemap.onehealthtrust.org/",
    "paper": null,
    "tags": [
      "AMR",
      "visualisation"
    ],
    "category": "infectious"
  },
  {
    "id": "vivli-amr",
    "name": "Vivli AMR Register",
    "description": "Industry AMR surveillance datasets (e.g. Pfizer ATLAS, GSK SOAR, J&J, Paratek) with millions of isolate-level susceptibility results.",
    "provider": "Vivli",
    "region": "Global",
    "access": "application",
    "access_note": "Free; registration and data request reviewed by the contributor.",
    "url": "https://amr.vivli.org/",
    "paper": null,
    "tags": [
      "AMR",
      "isolates"
    ],
    "category": "infectious"
  },
  {
    "id": "bv-brc",
    "name": "BV-BRC",
    "full_name": "Bacterial and Viral Bioinformatics Resource Center",
    "description": "Integrated genomes, AMR phenotypes, protein families and analysis tools for bacterial and viral pathogens (merger of PATRIC, IRD and ViPR).",
    "provider": "NIAID",
    "region": "Global",
    "access": "open",
    "access_note": "Open; free account to run analysis services.",
    "url": "https://www.bv-brc.org/",
    "paper": {
      "label": "Olson RD et al. Nucleic Acids Res 2023;51:D678",
      "url": "https://doi.org/10.1093/nar/gkac1003"
    },
    "tags": [
      "pathogen genomics",
      "AMR"
    ],
    "category": "infectious"
  },
  {
    "id": "ncbi-pathogen",
    "name": "NCBI Pathogen Detection",
    "description": "Real-time clustering of bacterial and fungal pathogen genomes from food, environment and patients, with AMR gene calls (AMRFinderPlus).",
    "provider": "NCBI",
    "region": "Global",
    "access": "open",
    "access_note": "Open.",
    "url": "https://www.ncbi.nlm.nih.gov/pathogens/",
    "paper": null,
    "tags": [
      "pathogen genomics",
      "outbreaks",
      "AMR"
    ],
    "category": "infectious"
  },
  {
    "id": "ncbi-virus",
    "name": "NCBI Virus",
    "description": "Portal for searching and downloading viral sequences (incl. SARS-CoV-2, influenza, dengue, mpox) with metadata.",
    "provider": "NCBI",
    "region": "Global",
    "access": "open",
    "access_note": "Open.",
    "url": "https://www.ncbi.nlm.nih.gov/labs/virus/",
    "paper": null,
    "tags": [
      "viral genomics"
    ],
    "category": "infectious"
  },
  {
    "id": "gisaid",
    "name": "GISAID",
    "description": "Global initiative sharing influenza, SARS-CoV-2, RSV, mpox and other virus genomes with associated epidemiological metadata.",
    "provider": "GISAID Initiative",
    "region": "Global",
    "access": "registration",
    "access_note": "Free; registration with identity verification and acceptance of the Database Access Agreement.",
    "url": "https://gisaid.org/",
    "paper": {
      "label": "Shu Y, McCauley J. Euro Surveill 2017;22:30494",
      "url": "https://doi.org/10.2807/1560-7917.ES.2017.22.13.30494"
    },
    "tags": [
      "viral genomics",
      "influenza",
      "SARS-CoV-2"
    ],
    "category": "infectious"
  },
  {
    "id": "nextstrain",
    "name": "Nextstrain",
    "description": "Real-time phylogenetic tracking of pathogen evolution (SARS-CoV-2, influenza, mpox, dengue, measles…) with open curated datasets and tools.",
    "provider": "Fred Hutch / University of Basel",
    "region": "Global",
    "access": "open",
    "access_note": "Open.",
    "url": "https://nextstrain.org/",
    "paper": {
      "label": "Hadfield J et al. Bioinformatics 2018;34:4121",
      "url": "https://doi.org/10.1093/bioinformatics/bty407"
    },
    "tags": [
      "phylodynamics",
      "genomic epidemiology"
    ],
    "category": "infectious"
  },
  {
    "id": "project-tycho",
    "name": "Project Tycho",
    "description": "Digitised US notifiable-disease reports since 1888 and global disease counts standardised for research on historical epidemics and vaccination impact.",
    "provider": "University of Pittsburgh",
    "region": "USA / Global",
    "access": "registration",
    "access_note": "Free registration.",
    "url": "https://www.tycho.pitt.edu/",
    "paper": {
      "label": "van Panhuis WG et al. N Engl J Med 2013;369:2152",
      "url": "https://doi.org/10.1056/NEJMms1215400"
    },
    "tags": [
      "historical",
      "notifiable diseases"
    ],
    "category": "infectious"
  },
  {
    "id": "delphi-epidata",
    "name": "Delphi Epidata (COVIDcast)",
    "description": "API of real-time US epidemic indicators — cases, hospitalisations, claims, symptom surveys, influenza-like illness — from Carnegie Mellon's Delphi group.",
    "provider": "Carnegie Mellon University",
    "region": "USA",
    "access": "open",
    "access_note": "Open API (free key recommended).",
    "url": "https://delphi.cmu.edu/",
    "paper": {
      "label": "Reinhart A et al. PNAS 2021;118:e2111452118",
      "url": "https://doi.org/10.1073/pnas.2111452118"
    },
    "tags": [
      "COVID-19",
      "influenza",
      "API"
    ],
    "category": "infectious"
  },
  {
    "id": "jhu-csse",
    "name": "JHU CSSE COVID-19 Data (archived)",
    "description": "The iconic global daily COVID-19 case and death time series (Jan 2020–Mar 2023) that powered the Johns Hopkins dashboard.",
    "provider": "Johns Hopkins University",
    "region": "Global",
    "access": "open",
    "access_note": "Direct download from GitHub (CC BY 4.0); no longer updated.",
    "url": "https://github.com/CSSEGISandData/COVID-19",
    "paper": {
      "label": "Dong E et al. Lancet Infect Dis 2020;20:533",
      "url": "https://doi.org/10.1016/S1473-3099(20)30120-1"
    },
    "tags": [
      "COVID-19",
      "time series"
    ],
    "category": "infectious"
  },
  {
    "id": "ecdc-atlas",
    "name": "ECDC Surveillance Atlas of Infectious Diseases",
    "description": "Interactive EU/EEA surveillance data for ~50 communicable diseases including AMR (EARS-Net), with downloadable tables.",
    "provider": "European Centre for Disease Prevention and Control",
    "region": "Europe",
    "access": "open",
    "access_note": "Open.",
    "url": "https://atlas.ecdc.europa.eu/",
    "paper": null,
    "tags": [
      "surveillance",
      "Europe",
      "AMR"
    ],
    "category": "infectious"
  },
  {
    "id": "global-health-linelist",
    "name": "Global.health",
    "description": "Open, de-identified line-list case data for outbreaks (COVID-19, mpox, Marburg, H5N1) curated from public sources.",
    "provider": "Global.health consortium (Oxford, Harvard, Boston Children's)",
    "region": "Global",
    "access": "open",
    "access_note": "Open download (some datasets require free login).",
    "url": "https://global.health/",
    "paper": null,
    "tags": [
      "outbreaks",
      "line lists"
    ],
    "category": "infectious"
  },
  {
    "id": "flusense",
    "name": "FluSense",
    "description": "350,000+ waiting-room thermal images and 21 million audio samples (coughs, speech) from a university clinic for influenza-like-illness surveillance.",
    "provider": "UMass Amherst",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from GitHub.",
    "url": "https://github.com/Forsad/FluSense-data",
    "paper": {
      "label": "Al Hossain F et al. Proc ACM IMWUT 2020;4:1",
      "url": "https://doi.org/10.1145/3381014"
    },
    "tags": [
      "influenza",
      "audio",
      "surveillance"
    ],
    "category": "infectious"
  },
  {
    "id": "protective-policy-index",
    "name": "Protective Policy Index (PPI)",
    "description": "Daily index of COVID-19 protective policy stringency for US states and other jurisdictions during 2020.",
    "provider": "COVID-19 Policy Response Lab",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from GitHub.",
    "url": "https://github.com/COVID-policy-response-lab/PPI-data",
    "paper": null,
    "tags": [
      "COVID-19",
      "policy"
    ],
    "category": "infectious"
  },
  {
    "id": "physionet",
    "name": "PhysioNet",
    "description": "The home of MIMIC, eICU and hundreds of open and credentialed physiological, clinical and imaging datasets, plus annual Computing in Cardiology challenges.",
    "provider": "MIT Laboratory for Computational Physiology",
    "region": "International",
    "access": "open",
    "access_note": "Mixed: each dataset is Open, Restricted (registration + DUA) or Credentialed.",
    "url": "https://physionet.org/",
    "paper": {
      "label": "Goldberger AL et al. Circulation 2000;101:e215",
      "url": "https://doi.org/10.1161/01.CIR.101.23.e215"
    },
    "tags": [
      "hub",
      "physiological signals",
      "EHR"
    ],
    "category": "hubs"
  },
  {
    "id": "uci-ml",
    "name": "UCI Machine Learning Repository",
    "description": "Long-running repository of 600+ curated datasets, including dozens of classic clinical ones (breast cancer Wisconsin, heart disease, diabetes, CKD, hepatitis).",
    "provider": "University of California, Irvine",
    "region": "International",
    "access": "open",
    "access_note": "Direct download (mostly CC BY 4.0).",
    "url": "https://archive.ics.uci.edu/",
    "paper": null,
    "tags": [
      "hub",
      "teaching",
      "tabular"
    ],
    "category": "hubs"
  },
  {
    "id": "grand-challenge",
    "name": "Grand Challenge",
    "description": "Platform hosting hundreds of biomedical image-analysis challenges with downloadable datasets and leaderboard evaluation.",
    "provider": "Radboud UMC (DIAG)",
    "region": "International",
    "access": "registration",
    "access_note": "Free account; each challenge sets its own data terms.",
    "url": "https://grand-challenge.org/challenges/",
    "paper": null,
    "tags": [
      "hub",
      "imaging",
      "challenges"
    ],
    "category": "hubs"
  },
  {
    "id": "synapse",
    "name": "Synapse",
    "description": "Collaborative research platform hosting data for DREAM challenges, AMP-AD, BraTS and many consortia, with fine-grained access controls.",
    "provider": "Sage Bionetworks",
    "region": "International",
    "access": "registration",
    "access_note": "Free account; some datasets need certified-user quiz or data-use approval.",
    "url": "https://www.synapse.org/",
    "paper": null,
    "tags": [
      "hub",
      "challenges",
      "omics"
    ],
    "category": "hubs"
  },
  {
    "id": "kaggle",
    "name": "Kaggle Datasets",
    "description": "Community data platform with thousands of medical datasets and competition data (RSNA, SIIM, APTOS, PANDA) plus free GPU notebooks.",
    "provider": "Google",
    "region": "International",
    "access": "registration",
    "access_note": "Free account; licences vary — check provenance carefully.",
    "url": "https://www.kaggle.com/datasets?tags=4202-Health",
    "paper": null,
    "tags": [
      "hub",
      "competitions"
    ],
    "category": "hubs"
  },
  {
    "id": "icpsr",
    "name": "ICPSR",
    "description": "The world's largest archive of social and behavioural science data, including health and ageing collections (NACDA) and many NIH-funded studies.",
    "provider": "University of Michigan",
    "region": "International",
    "access": "registration",
    "access_note": "Free account for public-use files; restricted files by application.",
    "url": "https://www.icpsr.umich.edu/",
    "paper": null,
    "tags": [
      "hub",
      "surveys",
      "ageing"
    ],
    "category": "hubs"
  },
  {
    "id": "ukds",
    "name": "UK Data Service",
    "description": "UK's largest collection of social, economic and population data — including ELSA, Understanding Society, Health Survey for England and birth cohorts.",
    "provider": "University of Essex / UKRI",
    "region": "United Kingdom",
    "access": "registration",
    "access_note": "Free registration (international users welcome for most safeguarded data).",
    "url": "https://ukdataservice.ac.uk/",
    "paper": null,
    "tags": [
      "hub",
      "surveys",
      "cohorts"
    ],
    "category": "hubs"
  },
  {
    "id": "aws-open-data",
    "name": "Registry of Open Data on AWS (life sciences)",
    "description": "Cloud-hosted public datasets you can analyse in place — e.g. gnomAD, 1000 Genomes, TCGA open tier, CAMELYON, EMBED, NIH CXR.",
    "provider": "Amazon Web Services",
    "region": "International",
    "access": "open",
    "access_note": "Mostly open; some entries require approval from the data owner.",
    "url": "https://registry.opendata.aws/?search=tags:life%20sciences",
    "paper": null,
    "tags": [
      "hub",
      "cloud"
    ],
    "category": "hubs"
  },
  {
    "id": "zenodo",
    "name": "Zenodo",
    "description": "CERN-run general-purpose open repository where many medical datasets (TotalSegmentator, CODE-15%, COUGHVID…) are deposited with DOIs.",
    "provider": "CERN / OpenAIRE",
    "region": "International",
    "access": "open",
    "access_note": "Open; per-record licences and occasional restricted access.",
    "url": "https://zenodo.org/",
    "paper": null,
    "tags": [
      "hub",
      "DOI"
    ],
    "category": "hubs"
  },
  {
    "id": "meddatasets-py",
    "name": "meddatasets (Python package)",
    "description": "Python package bundling curated medical and public-health datasets (cancer, chronic disease, hospital management, COVID-19, environmental health) for teaching and analysis.",
    "provider": "Renzo Caceres Rossi",
    "region": "International",
    "access": "open",
    "access_note": "pip install meddatasets (MIT licence); source on GitHub.",
    "url": "https://pypi.org/project/meddatasets/",
    "paper": null,
    "tags": [
      "Python package",
      "teaching"
    ],
    "category": "hubs"
  },
  {
    "id": "huggingface-datasets",
    "name": "Hugging Face Datasets",
    "description": "The largest hub for ML datasets, including thousands of medical text, imaging and benchmark datasets with one-line loaders.",
    "provider": "Hugging Face",
    "region": "International",
    "access": "registration",
    "access_note": "Most datasets open; gated ones need a free account and accepting terms.",
    "url": "https://huggingface.co/datasets?search=medical",
    "paper": null,
    "tags": [
      "hub",
      "AI"
    ],
    "category": "hubs"
  },
  {
    "id": "harvard-dataverse",
    "name": "Harvard Dataverse",
    "description": "Open research data repository with thousands of medicine, health and life-science datasets deposited with DOIs.",
    "provider": "Harvard University",
    "region": "International",
    "access": "open",
    "access_note": "Mostly open; some datasets restricted by depositors.",
    "url": "https://dataverse.harvard.edu/",
    "paper": null,
    "tags": [
      "hub",
      "DOI"
    ],
    "category": "hubs"
  },
  {
    "id": "mendeley-data",
    "name": "Mendeley Data",
    "description": "Generalist repository hosting many published clinical and biomedical datasets (e.g. Kermany OCT, PAD-UFES-20).",
    "provider": "Elsevier",
    "region": "International",
    "access": "open",
    "access_note": "Open; per-dataset licences.",
    "url": "https://data.mendeley.com/",
    "paper": null,
    "tags": [
      "hub",
      "DOI"
    ],
    "category": "hubs"
  },
  {
    "id": "healthbench-professional",
    "name": "HealthBench Professional",
    "description": "525 physician-authored clinician tasks with rubrics for evaluating how well LLMs support professional clinical work.",
    "provider": "OpenAI",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from Hugging Face.",
    "url": "https://huggingface.co/datasets/openai/healthbench-professional",
    "paper": null,
    "tags": [
      "LLM evaluation",
      "rubrics",
      "clinician tasks"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "medval-bench",
    "name": "MedVAL-Bench",
    "description": "840 AI-generated medical texts (reports, summaries, notes) graded by physicians for factual errors and patient-safety risk.",
    "provider": "Stanford MIMI",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from Hugging Face.",
    "url": "https://huggingface.co/datasets/stanfordmimi/MedVAL-Bench",
    "paper": null,
    "tags": [
      "LLM evaluation",
      "safety",
      "validation"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "vita-healthbench",
    "name": "VITA HealthBench datasets",
    "description": "Questions, model responses and rubric scores used to benchmark health AI systems: 500 questions across five systems plus 4,023 English-language questions.",
    "provider": "VITA project",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from figshare.",
    "url": "https://figshare.com/articles/dataset/VITA_Benchmarking_Questions_and_Output_Dataset/33224340",
    "paper": null,
    "tags": [
      "LLM evaluation",
      "health QA"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "medagentbench",
    "name": "MedAgentBench",
    "description": "300 physician-written clinical tasks in a simulated FHIR-based EHR environment for benchmarking LLM agents that act on patient records.",
    "provider": "Stanford University",
    "region": "USA",
    "access": "open",
    "access_note": "Code and data on GitHub.",
    "url": "https://github.com/stanfordmlgroup/MedAgentBench",
    "paper": {
      "label": "Jiang Y et al. arXiv 2025",
      "url": "https://arxiv.org/abs/2501.14654"
    },
    "tags": [
      "LLM agents",
      "FHIR",
      "EHR"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "medalign",
    "name": "MedAlign",
    "description": "983 clinician-written instructions over 276 longitudinal EHRs with 303 reference responses, for evaluating LLMs on real EHR tasks.",
    "provider": "Stanford Shah Lab",
    "region": "USA",
    "access": "application",
    "access_note": "Instructions public; EHR data via Stanford Redivis with a research data use agreement.",
    "url": "https://github.com/som-shahlab/medalign/",
    "paper": {
      "label": "Fleming SL et al. AAAI 2024",
      "url": "https://arxiv.org/abs/2308.14089"
    },
    "tags": [
      "LLM evaluation",
      "EHR"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "ehrshot",
    "name": "EHRSHOT",
    "description": "Longitudinal structured EHR data for 6,739 Stanford patients with 15 few-shot prediction tasks, released with a pretrained foundation model.",
    "provider": "Stanford Shah Lab",
    "region": "USA",
    "access": "application",
    "access_note": "Free research access via Stanford Redivis after accepting the data use agreement.",
    "url": "https://som-shahlab.github.io/ehrshot-website/",
    "paper": {
      "label": "Wornow M et al. NeurIPS 2023 Datasets & Benchmarks",
      "url": "https://arxiv.org/abs/2307.02028"
    },
    "tags": [
      "EHR",
      "few-shot",
      "foundation models"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "mimic-iv-ext-medicalbench",
    "name": "MIMIC-IV-Ext-MedicalBench",
    "description": "405 MIMIC-IV-derived examples for benchmarking extraction of ICD-10 diagnosis concepts from clinical text.",
    "provider": "PhysioNet",
    "region": "USA",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + CITI training + DUA.",
    "url": "https://physionet.org/content/mimic-iv-ext-medicalbench/",
    "paper": null,
    "tags": [
      "ICD coding",
      "MIMIC",
      "NLP"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "archehr-qa",
    "name": "ArchEHR-QA",
    "description": "Patient questions paired with clinician-rewritten questions and grounding EHR note excerpts, from the BioNLP 2025 shared task on answering patient questions from records.",
    "provider": "BioNLP / PhysioNet",
    "region": "USA",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + DUA.",
    "url": "https://physionet.org/content/archehr-qa-bionlp-task-2025/",
    "paper": null,
    "tags": [
      "QA",
      "patient questions",
      "EHR"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "verifact-bhc",
    "name": "VeriFact-BHC",
    "description": "13,070 propositions from human- and LLM-written brief hospital courses (MIMIC-III) annotated for factual support against the EHR.",
    "provider": "Stanford University / PhysioNet",
    "region": "USA",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + DUA.",
    "url": "https://physionet.org/content/mimic-iii-ext-verifact-bhc/",
    "paper": null,
    "tags": [
      "fact checking",
      "summarisation",
      "MIMIC"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "factehr",
    "name": "FactEHR",
    "description": "2,168 clinical notes from four types with LLM fact decompositions and entailment labels for evaluating factuality in clinical text.",
    "provider": "Stanford University",
    "region": "USA",
    "access": "credentialed",
    "access_note": "Built on MIMIC/PhysioNet notes; access through PhysioNet credentialing.",
    "url": "https://arxiv.org/abs/2412.12422",
    "paper": {
      "label": "Munnangi M et al. arXiv 2024",
      "url": "https://arxiv.org/abs/2412.12422"
    },
    "tags": [
      "factuality",
      "clinical notes"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "medec",
    "name": "MEDEC",
    "description": "3,848 clinical texts with injected or real medical errors for benchmarking error detection and correction by LLMs (MEDIQA-CORR 2024).",
    "provider": "Microsoft / University of Washington",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from GitHub.",
    "url": "https://github.com/abachaa/MEDEC",
    "paper": {
      "label": "Ben Abacha A et al. arXiv 2024",
      "url": "https://arxiv.org/abs/2412.19260"
    },
    "tags": [
      "error detection",
      "clinical notes"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "aci-bench",
    "name": "ACI-Bench",
    "description": "Ambient Clinical Intelligence benchmark: 207 doctor–patient conversation transcripts paired with visit notes for note-generation research.",
    "provider": "Microsoft / University of Washington",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from figshare / GitHub (CC BY 4.0).",
    "url": "https://github.com/wyim/aci-bench",
    "paper": {
      "label": "Yim W et al. Sci Data 2023;10:586",
      "url": "https://doi.org/10.1038/s41597-023-02487-3"
    },
    "tags": [
      "note generation",
      "dialogue"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "mts-dialog",
    "name": "MTS-Dialog",
    "description": "1,700 short doctor–patient conversations with corresponding summaries (clinical note sections) for dialogue summarisation.",
    "provider": "Microsoft / Ben Abacha et al.",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from GitHub.",
    "url": "https://github.com/abachaa/MTS-Dialog",
    "paper": {
      "label": "Ben Abacha A et al. EACL 2023",
      "url": "https://doi.org/10.18653/v1/2023.eacl-main.168"
    },
    "tags": [
      "dialogue",
      "summarisation"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "primock57",
    "name": "PriMock57",
    "description": "57 mock primary-care consultations with audio, transcripts and clinician-written notes for speech recognition and note generation.",
    "provider": "Babylon Health",
    "region": "United Kingdom",
    "access": "open",
    "access_note": "Direct download from GitHub.",
    "url": "https://github.com/babylonhealth/primock57",
    "paper": {
      "label": "Korfiatis AP et al. ACL 2022",
      "url": "https://doi.org/10.18653/v1/2022.acl-short.65"
    },
    "tags": [
      "audio",
      "dialogue",
      "note generation"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "pmc-patients",
    "name": "PMC-Patients",
    "description": "~167,000 patient summaries extracted from PubMed Central case reports with patient–article and patient–patient relevance labels for retrieval tasks.",
    "provider": "Tsinghua University",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from figshare / Hugging Face.",
    "url": "https://figshare.com/collections/PMC-Patients/6723465",
    "paper": {
      "label": "Zhao Z et al. Sci Data 2023;10:909",
      "url": "https://doi.org/10.1038/s41597-023-02814-8"
    },
    "tags": [
      "case reports",
      "retrieval"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "multicare",
    "name": "MultiCaRe",
    "description": "Multimodal case-report dataset from PubMed Central: 75,382 articles, 96,428 cases and 135,596 images with captions and labels.",
    "provider": "Nievas Offidani M, Delrieux C",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from Zenodo (CC BY).",
    "url": "https://zenodo.org/records/10079370",
    "paper": null,
    "tags": [
      "case reports",
      "images",
      "multimodal"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "nhse-synthetic-notes",
    "name": "NHS England Synthetic Clinical Notes",
    "description": "Synthetic clinical notes for 70 fictional patients, released by NHS England for safe NLP development.",
    "provider": "NHS England Data Science",
    "region": "United Kingdom",
    "access": "open",
    "access_note": "Direct download from Hugging Face.",
    "url": "https://huggingface.co/datasets/NHSEDataScience/synthetic_clinical_notes",
    "paper": null,
    "tags": [
      "synthetic",
      "clinical notes"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "ekacare-datasets",
    "name": "EkaCare Medical Public Datasets",
    "description": "Collection of Indian clinical datasets for medical speech recognition, health-record parsing and clinical note generation.",
    "provider": "Eka Care",
    "region": "India",
    "access": "open",
    "access_note": "Hugging Face collection; some sets require accepting terms.",
    "url": "https://huggingface.co/collections/ekacare/ekacare-medical-public-datasets",
    "paper": null,
    "tags": [
      "ASR",
      "South Asia",
      "records parsing"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "cardiac-investigations-text",
    "name": "Cardiac Investigations Text (AMIE)",
    "description": "Cardiology test reports (ECG, echo, Holter, CMR, CPET) released with Google's AMIE cardiology study.",
    "provider": "Google / Stanford",
    "region": "USA",
    "access": "application",
    "access_note": "Free; request access on Redivis.",
    "url": "https://redivis.com/datasets/1z3x-2354972da",
    "paper": null,
    "tags": [
      "cardiology",
      "reports",
      "LLM evaluation"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "parrot-reports",
    "name": "PARROT",
    "description": "2,658 fictional radiology reports written by radiologists in 14 languages, for multilingual radiology NLP.",
    "provider": "PARROT consortium",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from GitHub.",
    "url": "https://github.com/PARROT-reports/PARROT_v1.0",
    "paper": null,
    "tags": [
      "radiology reports",
      "multilingual"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "radgraph",
    "name": "RadGraph & RadGraph2",
    "description": "Radiologist-annotated entities and relations in chest X-ray reports (RadGraph: 500 reports; RadGraph2 adds change-over-time relations), plus large auto-annotated sets.",
    "provider": "Stanford University / PhysioNet",
    "region": "USA",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + DUA.",
    "url": "https://physionet.org/content/radgraph/",
    "paper": {
      "label": "Jain S et al. NeurIPS 2021 Datasets & Benchmarks",
      "url": "https://arxiv.org/abs/2106.14463"
    },
    "tags": [
      "radiology reports",
      "information extraction"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "radevalx",
    "name": "RadEvalX",
    "description": "Radiologist annotations of clinically significant errors in 100 AI-generated chest X-ray reports.",
    "provider": "PhysioNet",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from PhysioNet.",
    "url": "https://physionet.org/content/rad-eval-x/",
    "paper": null,
    "tags": [
      "report generation",
      "evaluation"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "rexerr",
    "name": "ReXErr-v1",
    "description": "Clinically meaningful synthetic errors injected into MIMIC-CXR reports for training and evaluating report-error detection.",
    "provider": "Harvard Rajpurkar Lab",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from PhysioNet.",
    "url": "https://physionet.org/content/rexerr-v1/",
    "paper": null,
    "tags": [
      "radiology reports",
      "error detection"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "chest-imagenome",
    "name": "Chest ImaGenome",
    "description": "Scene graphs linking anatomy, findings and comparisons for MIMIC-CXR images, with a 500-patient gold standard.",
    "provider": "IBM / MIT / PhysioNet",
    "region": "USA",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + DUA.",
    "url": "https://physionet.org/content/chest-imagenome/",
    "paper": null,
    "tags": [
      "chest X-ray",
      "scene graphs"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "reflacx",
    "name": "REFLACX",
    "description": "Eye-tracking data, dictated reports and abnormality ellipses from radiologists reading 3,032 MIMIC-CXR chest X-rays.",
    "provider": "University of Utah / PhysioNet",
    "region": "USA",
    "access": "registration",
    "access_note": "Free PhysioNet account + signed data use agreement.",
    "url": "https://physionet.org/content/reflacx-xray-localization/",
    "paper": {
      "label": "Bigolin Lanfredi R et al. Sci Data 2022;9:350",
      "url": "https://doi.org/10.1038/s41597-022-01441-z"
    },
    "tags": [
      "eye tracking",
      "chest X-ray"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "egd-cxr",
    "name": "Eye Gaze Data for Chest X-rays (EGD-CXR)",
    "description": "Eye-tracking and report dictation from a radiologist reading 1,083 MIMIC-CXR images.",
    "provider": "IBM Research / PhysioNet",
    "region": "USA",
    "access": "credentialed",
    "access_note": "PhysioNet credentialing + DUA.",
    "url": "https://physionet.org/content/egd-cxr/",
    "paper": {
      "label": "Karargyris A et al. Sci Data 2021;8:92",
      "url": "https://doi.org/10.1038/s41597-021-00863-5"
    },
    "tags": [
      "eye tracking",
      "chest X-ray"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "chexthought",
    "name": "CheXthought",
    "description": "Radiologist think-aloud reasoning traces and eye-tracking recorded while interpreting chest X-rays.",
    "provider": "Stanford University",
    "region": "USA",
    "access": "application",
    "access_note": "Free; request access on Stanford Redivis.",
    "url": "https://stanford.redivis.com/datasets/ds7j-75krm2btv",
    "paper": null,
    "tags": [
      "reasoning",
      "eye tracking",
      "chest X-ray"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "collab-cxr",
    "name": "Collab-CXR",
    "description": "Experimental data from 227 radiologists reading chest X-rays with and without AI assistance, for studying human–AI collaboration.",
    "provider": "MIT / Harvard",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from OSF.",
    "url": "https://osf.io/z7apq/",
    "paper": null,
    "tags": [
      "human-AI collaboration",
      "chest X-ray"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "llava-rad",
    "name": "LLaVA-Rad training data",
    "description": "697,000 radiology image–text pairs (400,042 chest X-ray pairs) assembled from public sources to train a small multimodal radiology model.",
    "provider": "Microsoft Research",
    "region": "International",
    "access": "credentialed",
    "access_note": "Annotations on GitHub; MIMIC-CXR images need PhysioNet credentialing.",
    "url": "https://github.com/microsoft/LLaVA-Rad",
    "paper": {
      "label": "Zambrano Chaves JM et al. arXiv 2024",
      "url": "https://arxiv.org/abs/2403.08002"
    },
    "tags": [
      "vision-language",
      "radiology"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "vqa-rad",
    "name": "VQA-RAD",
    "description": "3,515 clinician-generated question–answer pairs about 315 radiology images, the first manually built medical VQA set.",
    "provider": "NIH / Lau et al.",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from OSF.",
    "url": "https://osf.io/89kps/",
    "paper": {
      "label": "Lau JJ et al. Sci Data 2018;5:180251",
      "url": "https://doi.org/10.1038/sdata.2018.251"
    },
    "tags": [
      "VQA",
      "radiology"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "openmedreason",
    "name": "OpenMedReason",
    "description": "Medical visual question answering with image-grounded reasoning traces and a rubric-scored test split.",
    "provider": "Negin B. et al.",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from Hugging Face.",
    "url": "https://huggingface.co/datasets/neginb/OpenMedReason",
    "paper": null,
    "tags": [
      "VQA",
      "reasoning"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "woundcarevqa",
    "name": "WoundcareVQA",
    "description": "Multilingual wound-care visual question answering: 477 patient cases with images, queries and expert responses (MEDIQA-WV).",
    "provider": "MEDIQA organisers",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from OSF.",
    "url": "https://osf.io/xsj5u/",
    "paper": null,
    "tags": [
      "VQA",
      "wound care",
      "multilingual"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "synthvision",
    "name": "SynthVision",
    "description": "119,000 synthetic medical images with VQA annotations for training multimodal models without patient data.",
    "provider": "OpenMed Labs",
    "region": "International",
    "access": "open",
    "access_note": "Direct download via GitHub / Hugging Face.",
    "url": "https://github.com/openmed-labs/synthvision",
    "paper": null,
    "tags": [
      "synthetic",
      "VQA"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "biomedica",
    "name": "BIOMEDICA",
    "description": "24 million image–caption pairs from 6 million PubMed Central open-access articles for training biomedical vision-language models.",
    "provider": "Stanford University",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from Hugging Face.",
    "url": "https://huggingface.co/BIOMEDICA",
    "paper": {
      "label": "Lozano A et al. arXiv 2025",
      "url": "https://arxiv.org/abs/2501.07171"
    },
    "tags": [
      "vision-language",
      "literature"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "biomedparsedata",
    "name": "BiomedParseData",
    "description": "3.4 million image–mask–label triples across nine imaging modalities used to train the BiomedParse segmentation foundation model.",
    "provider": "Microsoft Research",
    "region": "International",
    "access": "registration",
    "access_note": "Hugging Face account and acceptance of terms (gated).",
    "url": "https://huggingface.co/datasets/microsoft/BiomedParseData",
    "paper": {
      "label": "Zhao T et al. Nat Methods 2025;22:166",
      "url": "https://doi.org/10.1038/s41592-024-02499-w"
    },
    "tags": [
      "segmentation",
      "multimodal",
      "foundation models"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "openpath",
    "name": "OpenPath (PLIP)",
    "description": "208,414 pathology images paired with natural-language descriptions curated from medical Twitter, used to train the PLIP model.",
    "provider": "Stanford University",
    "region": "International",
    "access": "registration",
    "access_note": "Tweet IDs and links released; images retrieved per the authors' instructions.",
    "url": "https://huggingface.co/spaces/vinid/webplip",
    "paper": {
      "label": "Huang Z et al. Nat Med 2023;29:2307",
      "url": "https://doi.org/10.1038/s41591-023-02504-3"
    },
    "tags": [
      "pathology",
      "vision-language"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "ecg-grounding",
    "name": "ECG-Grounding",
    "description": "30,000 ECG instruction pairs with grounded, heartbeat-level interpretations for training explainable ECG language models.",
    "provider": "LANSG",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from Hugging Face.",
    "url": "https://huggingface.co/datasets/LANSG/ECG-Grounding",
    "paper": null,
    "tags": [
      "ECG",
      "instruction tuning"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "openbiomedvid",
    "name": "OpenBiomedVid",
    "description": "1,031 hours of biomedical video–text instruction data mined from educational videos, with SurgeryVideoQA and MIMICEchoQA benchmarks.",
    "provider": "Stanford Zou Group",
    "region": "International",
    "access": "open",
    "access_note": "Code and data links on GitHub.",
    "url": "https://github.com/zou-group/OpenBiomedVid",
    "paper": null,
    "tags": [
      "video",
      "vision-language"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "medvidbench",
    "name": "MedVidBench",
    "description": "6,245 test samples benchmarking video-language models on surgical and medical video understanding.",
    "provider": "United Imaging Intelligence",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from Hugging Face.",
    "url": "https://huggingface.co/datasets/UII-AI/MedVidBench",
    "paper": null,
    "tags": [
      "video",
      "benchmark",
      "surgery"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "medconceptsqa",
    "name": "MedConceptsQA",
    "description": "Open benchmark of questions on medical concepts (ICD-9/10, ATC and procedure codes) at three difficulty levels.",
    "provider": "Shoham OB, Rappoport N",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from Hugging Face.",
    "url": "https://huggingface.co/datasets/ofir408/MedConceptsQA",
    "paper": {
      "label": "Shoham OB, Rappoport N. arXiv 2024",
      "url": "https://arxiv.org/abs/2405.07348"
    },
    "tags": [
      "QA",
      "medical codes"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "equitymedqa",
    "name": "EquityMedQA",
    "description": "Seven adversarial question sets for surfacing health-equity-related biases and harms in medical LLM answers.",
    "provider": "Google Research",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from Hugging Face.",
    "url": "https://huggingface.co/datasets/katielink/EquityMedQA",
    "paper": {
      "label": "Pfohl SR et al. Nat Med 2024;30:3590",
      "url": "https://doi.org/10.1038/s41591-024-03258-2"
    },
    "tags": [
      "bias",
      "equity",
      "LLM evaluation"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "med-halt",
    "name": "Med-HALT",
    "description": "Benchmark of reasoning and memory tests built from multinational medical exams for measuring hallucinations in LLMs.",
    "provider": "Saama AI Research",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from Hugging Face.",
    "url": "https://huggingface.co/datasets/MedHALT/Med-HALT",
    "paper": {
      "label": "Pal A et al. CoNLL 2023",
      "url": "https://doi.org/10.18653/v1/2023.conll-1.21"
    },
    "tags": [
      "hallucination",
      "LLM evaluation"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "red-teaming-llm-medicine",
    "name": "Red Teaming LLMs in Medicine",
    "description": "382 clinician-crafted adversarial prompts with 1,146 ChatGPT responses labelled for safety, privacy, hallucination and bias.",
    "provider": "Stanford Daneshjou Lab",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from project site.",
    "url": "https://daneshjoulab.github.io/Red-Teaming-Dataset/",
    "paper": null,
    "tags": [
      "safety",
      "red teaming"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "q-pain",
    "name": "Q-Pain",
    "description": "Clinical vignettes on pain management designed to measure social bias (race, gender) in medical QA systems.",
    "provider": "Stanford / PhysioNet",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from PhysioNet.",
    "url": "https://physionet.org/content/q-pain/",
    "paper": null,
    "tags": [
      "bias",
      "QA"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "coral",
    "name": "CORAL",
    "description": "Expert-annotated oncology progress notes (breast and pancreatic cancer) for evaluating LLM information extraction.",
    "provider": "UCSF",
    "region": "USA",
    "access": "credentialed",
    "access_note": "Distributed via PhysioNet with credentialing + DUA.",
    "url": "https://physionet.org/content/curated-oncology-reports/",
    "paper": {
      "label": "Sushil M et al. NEJM AI 2024",
      "url": "https://doi.org/10.1056/AIdbp2300110"
    },
    "tags": [
      "oncology",
      "information extraction"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "meqsum",
    "name": "MeQSum",
    "description": "1,000 consumer health questions paired with expert summaries for question summarisation.",
    "provider": "NLM / Ben Abacha et al.",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from GitHub / Hugging Face.",
    "url": "https://github.com/abachaa/MeQSum",
    "paper": {
      "label": "Ben Abacha A, Demner-Fushman D. ACL 2019",
      "url": "https://doi.org/10.18653/v1/P19-1215"
    },
    "tags": [
      "summarisation",
      "consumer health"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "emrqa",
    "name": "emrQA",
    "description": "~1 million question–logical form and 400,000 question–answer pairs generated from n2c2 clinical notes for EMR question answering.",
    "provider": "IBM Research / UIUC",
    "region": "USA",
    "access": "application",
    "access_note": "Generation scripts on GitHub; underlying notes require n2c2 DUA.",
    "url": "https://github.com/panushri25/emrQA",
    "paper": {
      "label": "Pampari A et al. EMNLP 2018",
      "url": "https://doi.org/10.18653/v1/D18-1258"
    },
    "tags": [
      "QA",
      "clinical notes"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "codiesp",
    "name": "CodiEsp",
    "description": "1,000 Spanish clinical case reports manually coded with ICD-10 diagnoses and procedures (CLEF eHealth 2020).",
    "provider": "Barcelona Supercomputing Center",
    "region": "Spain",
    "access": "open",
    "access_note": "Direct download from Zenodo (CC BY 4.0).",
    "url": "https://zenodo.org/records/3837305",
    "paper": null,
    "tags": [
      "ICD coding",
      "Spanish",
      "NLP"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "gptnermed",
    "name": "GPTNERMED",
    "description": "9,845 synthetic German medical sentences with named-entity annotations generated by GPT (not expert-validated).",
    "provider": "Frei J, Kramer F",
    "region": "Germany",
    "access": "open",
    "access_note": "Direct download from Hugging Face.",
    "url": "https://huggingface.co/datasets/jfrei/GPTNERMED",
    "paper": {
      "label": "Frei J, Kramer F. J Biomed Inform 2023;144:104478",
      "url": "https://doi.org/10.1016/j.jbi.2023.104478"
    },
    "tags": [
      "NER",
      "German",
      "synthetic"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "biored",
    "name": "BioRED",
    "description": "600 PubMed abstracts annotated for biomedical entities and eight relation types for relation extraction.",
    "provider": "NCBI / NLM",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from NCBI FTP.",
    "url": "https://ftp.ncbi.nlm.nih.gov/pub/lu/BioRED/",
    "paper": {
      "label": "Luo L et al. Brief Bioinform 2022;23:bbac282",
      "url": "https://doi.org/10.1093/bib/bbac282"
    },
    "tags": [
      "relation extraction",
      "literature"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "pgxcorpus",
    "name": "PGxCorpus",
    "description": "945 sentences from 911 PubMed abstracts manually annotated with pharmacogenomic entities and relations.",
    "provider": "LORIA / Inria",
    "region": "France",
    "access": "open",
    "access_note": "Direct download from figshare.",
    "url": "https://doi.org/10.6084/m9.figshare.6025748",
    "paper": {
      "label": "Legrand J et al. Sci Data 2020;7:3",
      "url": "https://doi.org/10.1038/s41597-019-0342-9"
    },
    "tags": [
      "pharmacogenomics",
      "relation extraction"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "bigbio",
    "name": "BigBIO",
    "description": "Community library of 126+ biomedical NLP datasets in a uniform schema with Hugging Face loaders.",
    "provider": "BigScience workshop",
    "region": "International",
    "access": "open",
    "access_note": "Open; some included datasets carry their own DUAs.",
    "url": "https://github.com/bigscience-workshop/biomedical",
    "paper": {
      "label": "Fries JA et al. NeurIPS 2022 Datasets & Benchmarks",
      "url": "https://arxiv.org/abs/2206.15076"
    },
    "tags": [
      "NLP",
      "hub"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "bridge-open",
    "name": "BRIDGE-Open",
    "description": "Open-access subset of the BRIDGE benchmark: 55 multilingual clinical text tasks from real-world notes for evaluating LLMs.",
    "provider": "Y-Lab, Harvard",
    "region": "International",
    "access": "registration",
    "access_note": "Hugging Face account and acceptance of terms (gated).",
    "url": "https://huggingface.co/datasets/YLab-Open/BRIDGE-Open",
    "paper": null,
    "tags": [
      "LLM evaluation",
      "multilingual"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "medinstruct-52k",
    "name": "MedInstruct-52k",
    "description": "52,000 diverse medical instruction–response pairs generated with GPT-4 for instruction-tuning (AlpaCare).",
    "provider": "UC Santa Barbara",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from GitHub.",
    "url": "https://github.com/XZhang97666/AlpaCare",
    "paper": {
      "label": "Zhang X et al. arXiv 2023",
      "url": "https://arxiv.org/abs/2310.14558"
    },
    "tags": [
      "instruction tuning",
      "synthetic"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "ultramedical",
    "name": "UltraMedical",
    "description": "410,000 biomedical instructions with preference data, used to train the UltraMedical model family.",
    "provider": "Tsinghua University",
    "region": "International",
    "access": "open",
    "access_note": "Direct download via GitHub / Hugging Face.",
    "url": "https://github.com/TsinghuaC3I/UltraMedical",
    "paper": {
      "label": "Zhang K et al. NeurIPS 2024 Datasets & Benchmarks",
      "url": "https://arxiv.org/abs/2406.03949"
    },
    "tags": [
      "instruction tuning",
      "preference data"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "ii-medical-reasoning",
    "name": "II-Medical-Reasoning-SFT",
    "description": "~2.2 million medical reasoning samples for supervised fine-tuning of reasoning LLMs.",
    "provider": "Intelligent Internet",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from Hugging Face.",
    "url": "https://huggingface.co/datasets/Intelligent-Internet/II-Medical-Reasoning-SFT",
    "paper": null,
    "tags": [
      "reasoning",
      "instruction tuning"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "openmed-reasoning-sft",
    "name": "Medical-Reasoning-SFT-GPT-OSS-120B",
    "description": "~200,000 synthetic medical reasoning conversations generated with GPT-OSS-120B.",
    "provider": "OpenMed",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from Hugging Face.",
    "url": "https://huggingface.co/datasets/OpenMed/Medical-Reasoning-SFT-GPT-OSS-120B",
    "paper": null,
    "tags": [
      "reasoning",
      "synthetic"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "llama2-medtuned",
    "name": "Llama2-MedTuned-Instructions",
    "description": "~200,000 biomedical instruction samples (NER, RE, NLI, QA, classification) for instruction-tuning LLMs.",
    "provider": "NLPie Research, University of Oxford",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from Hugging Face.",
    "url": "https://huggingface.co/datasets/nlpie/Llama2-MedTuned-Instructions",
    "paper": null,
    "tags": [
      "instruction tuning",
      "NLP"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "mediflow",
    "name": "MediFlow",
    "description": "2.5 million synthetic clinical instruction rows spanning many clinical text tasks and output formats.",
    "provider": "Microsoft",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from Hugging Face.",
    "url": "https://huggingface.co/datasets/microsoft/mediflow",
    "paper": null,
    "tags": [
      "instruction tuning",
      "synthetic"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "hc4",
    "name": "HC4",
    "description": "Healthcare-focused pretraining corpus of ~65 billion tokens curated from web and literature sources.",
    "provider": "M42 Health",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from Hugging Face.",
    "url": "https://huggingface.co/datasets/m42-health/HC4",
    "paper": null,
    "tags": [
      "pretraining corpus"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "nhs-llm-opengpt",
    "name": "NHS-LLM / OpenGPT datasets",
    "description": "24,665 NHS-derived Q&A pairs, 2,354 conversations and 4,688 medical task pairs generated with the OpenGPT framework.",
    "provider": "CogStack / King's College London",
    "region": "United Kingdom",
    "access": "open",
    "access_note": "Direct download from GitHub.",
    "url": "https://github.com/CogStack/opengpt",
    "paper": null,
    "tags": [
      "QA",
      "conversations"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "dialogues-of-delivery",
    "name": "Dialogues of Delivery",
    "description": "Maternal-health questions and answers in four languages for evaluating multilingual health chatbots.",
    "provider": "Harvard Dataverse deposit",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from Harvard Dataverse.",
    "url": "https://doi.org/10.7910/DVN/JS3ILT",
    "paper": null,
    "tags": [
      "maternal health",
      "multilingual",
      "QA"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "sgir-clinical-trials",
    "name": "CSIRO Clinical Trials Search Collection",
    "description": "204,855 clinical trials with relevance judgements for patient-to-trial matching and search evaluation.",
    "provider": "CSIRO",
    "region": "Australia",
    "access": "open",
    "access_note": "Direct download from CSIRO Data Access Portal.",
    "url": "https://data.csiro.au/collection/csiro:17152",
    "paper": null,
    "tags": [
      "trial matching",
      "information retrieval"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "trec-biomedical",
    "name": "TREC Biomedical Tracks",
    "description": "Test collections from TREC Clinical Decision Support, Precision Medicine and Clinical Trials tracks (topics, documents, relevance judgements).",
    "provider": "NIST / TREC",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from track sites.",
    "url": "http://www.trec-cds.org/",
    "paper": null,
    "tags": [
      "information retrieval",
      "trial matching"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "cto-dataset",
    "name": "CTO (Clinical Trial Outcome) Dataset",
    "description": "~479,000 clinical trials with automatically derived weak outcome labels for trial-outcome prediction.",
    "provider": "UIUC Sunlab",
    "region": "International",
    "access": "open",
    "access_note": "Direct download from GitHub / Hugging Face.",
    "url": "https://github.com/chufangao/CTOD",
    "paper": null,
    "tags": [
      "clinical trials",
      "outcome prediction"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "ai-cad-database",
    "name": "FDA AI-CAD Device Database",
    "description": "Curated list of FDA clearances for AI-based image-interpretation (CAD) devices with evidence characteristics.",
    "provider": "Lotter Lab, Dana-Farber",
    "region": "USA",
    "access": "open",
    "access_note": "Direct download from GitHub.",
    "url": "https://github.com/lotterlab/ai_cad_database",
    "paper": null,
    "tags": [
      "regulatory",
      "medical devices"
    ],
    "category": "ai-benchmarks"
  },
  {
    "id": "kvasir",
    "name": "Kvasir family (Kvasir v2, Kvasir-SEG, Kvasir-Capsule, HyperKvasir)",
    "description": "Gastrointestinal endoscopy image and video collections from Norway: 8,000 labelled images (v2), polyp segmentation masks (SEG), 4.7M video-capsule frames (Capsule) and 110,000+ images (HyperKvasir).",
    "provider": "Simula Research Laboratory / Vestre Viken Hospital",
    "region": "Norway",
    "access": "open",
    "access_note": "Direct download for research and education.",
    "url": "https://datasets.simula.no/kvasir/",
    "paper": {
      "label": "Pogorelov K et al. ACM MMSys 2017",
      "url": "https://doi.org/10.1145/3083187.3083212"
    },
    "tags": [
      "endoscopy",
      "polyps",
      "capsule"
    ],
    "category": "surgery-endoscopy"
  },
  {
    "id": "nerthus",
    "name": "Nerthus",
    "description": "21 colonoscopy videos (5,525 frames) labelled with the Boston Bowel Preparation Scale for bowel-cleansing assessment.",
    "provider": "Simula Research Laboratory",
    "region": "Norway",
    "access": "open",
    "access_note": "Direct download.",
    "url": "https://datasets.simula.no/nerthus/",
    "paper": {
      "label": "Pogorelov K et al. ACM MMSys 2017",
      "url": "https://doi.org/10.1145/3083187.3083216"
    },
    "tags": [
      "colonoscopy",
      "bowel preparation"
    ],
    "category": "surgery-endoscopy"
  },
  {
    "id": "see-ai-capsule",
    "name": "SEE-AI Small Bowel Capsule Endoscopy",
    "description": "18,481 small-bowel capsule endoscopy images with bounding-box annotations for lesion detection.",
    "provider": "Kyushu University / SEE-AI project",
    "region": "Japan",
    "access": "registration",
    "access_note": "Free Kaggle account.",
    "url": "https://www.kaggle.com/datasets/capsuleyolo/kyucapsule",
    "paper": null,
    "tags": [
      "capsule endoscopy",
      "object detection"
    ],
    "category": "surgery-endoscopy"
  },
  {
    "id": "lemon-surgical",
    "name": "LEMON",
    "description": "4,000+ surgical videos (938 hours) across many procedure types for pretraining surgical video foundation models.",
    "provider": "visurg.ai / King's College London",
    "region": "International",
    "access": "open",
    "access_note": "Video list and download scripts on GitHub.",
    "url": "https://github.com/visurg-ai/LEMON",
    "paper": null,
    "tags": [
      "surgical video",
      "foundation models"
    ],
    "category": "surgery-endoscopy"
  },
  {
    "id": "surglavi",
    "name": "SurgLaVi",
    "description": "~240,000 surgical video clip–caption pairs for surgical vision-language pretraining; the open subset SurgLaVi-β is freely released.",
    "provider": "Pérez R. et al.",
    "region": "International",
    "access": "open",
    "access_note": "Open subset via GitHub.",
    "url": "https://github.com/aperezr20/SurgLaVi",
    "paper": null,
    "tags": [
      "surgical video",
      "vision-language"
    ],
    "category": "surgery-endoscopy"
  },
  {
    "id": "gr00t-h",
    "name": "GR00T-H Surgical Robot Data",
    "description": "778 hours of surgical-robot demonstration episodes released with NVIDIA's GR00T-H healthcare robotics model.",
    "provider": "NVIDIA",
    "region": "International",
    "access": "registration",
    "access_note": "Hugging Face account and acceptance of licence terms.",
    "url": "https://huggingface.co/nvidia/GR00T-H",
    "paper": null,
    "tags": [
      "surgical robotics",
      "demonstrations"
    ],
    "category": "surgery-endoscopy"
  },
  {
    "id": "hdr-gateway",
    "name": "Health Data Research Gateway",
    "description": "UK national catalogue of 1,400+ health datasets (NHS records, cohorts, biobanks, trials, registries) with standardised metadata, search filters and a common data-access request workflow.",
    "provider": "Health Data Research UK",
    "region": "United Kingdom",
    "access": "open",
    "access_note": "Catalogue free to search. Access to each dataset is via its custodian: many are free, some charge fees.",
    "url": "https://healthdatagateway.org/",
    "paper": null,
    "tags": [
      "hub",
      "catalogue",
      "UK"
    ],
    "category": "catalogues"
  },
  {
    "id": "brainxai-data",
    "name": "BrainX AI Data Catalogue",
    "description": "Community-curated list of ~300 medical AI datasets, benchmarks and resources (imaging, EHR, NLP, LLM evaluation, wearables) maintained by the BrainX AI community.",
    "provider": "BrainX Community",
    "region": "International",
    "access": "open",
    "access_note": "Free web list; each linked dataset has its own terms, and a few are fee-based.",
    "url": "https://brainxai.org/data/",
    "paper": null,
    "tags": [
      "catalogue",
      "AI",
      "benchmarks"
    ],
    "category": "catalogues"
  },
  {
    "id": "ghdx",
    "name": "Global Health Data Exchange (GHDx)",
    "description": "IHME's searchable catalogue of the world's surveys, censuses, vital registration and disease registries, linking to source data wherever available.",
    "provider": "Institute for Health Metrics and Evaluation",
    "region": "Global",
    "access": "open",
    "access_note": "Catalogue free to search; access terms of each linked dataset vary.",
    "url": "https://ghdx.healthdata.org/",
    "paper": null,
    "tags": [
      "catalogue",
      "metadata"
    ],
    "category": "catalogues"
  },
  {
    "id": "google-dataset-search",
    "name": "Google Dataset Search",
    "description": "Search engine indexing dataset descriptions across thousands of repositories — useful for finding datasets not listed here.",
    "provider": "Google",
    "region": "International",
    "access": "open",
    "access_note": "Free web search; results include both free and paid datasets.",
    "url": "https://datasetsearch.research.google.com/",
    "paper": null,
    "tags": [
      "search engine",
      "discovery"
    ],
    "category": "catalogues"
  },
  {
    "id": "mobilize-datasets",
    "name": "Mobilize Center Datasets",
    "description": "Curated list of wearable-sensor and human-movement datasets (accelerometry, biomechanics, gait) such as AddBiomechanics and NHANES accelerometry.",
    "provider": "Stanford Mobilize Center",
    "region": "International",
    "access": "open",
    "access_note": "Free list; each dataset has its own terms, and some (e.g. UK Biobank) are fee-based.",
    "url": "https://mobilize.stanford.edu/data/available-datasets/",
    "paper": null,
    "tags": [
      "catalogue",
      "wearables",
      "movement"
    ],
    "category": "catalogues"
  }
]