docs aireadi org docs-2 d4d

Datasheet for Dataset - Human Readable Format

🎯

Motivation

Why was the dataset created?

  • Response
    Advance AI/ML research to better understand salutogenesis (pathway from disease to health) in T2DM using harmonized, multimodal human subject data.
📊

Composition

What do the instances represent?

  1. Representation
    Individuals with and without Type 2 Diabetes Mellitus (T2DM)
    Instance Type
    Human participants and associated multimodal measurements collected across three clinical sites
    Data Type
    Multimodal observations per participant including survey data, physical measurements, clinical measurements (e.g., monofilament and cognitive function testing), retinal images, ECG, wearable device data, blood and urine laboratory results, blood glucose levels, and environmental variables (e.g., home air quality). Controlled-access components include genetic sequencing data and past health records.
    Sampling Strategies
    1. Is Sample
      • yes
      Is Random
      • no
      Source Data
      • Participants recruited at three clinical sites under a standardized protocol
      Is Representative
      • Recruitment aimed at approximately equal distribution across diabetes severity; actual pilot release(s) may not be balanced due to ongoing enrollment.
      Why Not Representative
      • Enrollment ongoing; pilot release and periodic updates may not achieve balance across groups.
      Strategies
      • Stratified recruitment targeting approximately equal distribution by diabetes severity
  • Identification
    • With T2DM vs. without T2DM
    • Diabetes severity strata
    Distribution
    • Recruitment targeted approximately equal distribution across diabetes severity; pilot release(s) may be imbalanced due to ongoing enrollment.
Description
Public subset downloadable from the FAIRhub data portal under a license agreement.
Full dataset available under a data use agreement (controlled access) via the FAIRhub data portal.
🔍

Collection Process

How was the data acquired?

ai-readi-flagship-t2dm
AI-READI Flagship Dataset of Type 2 Diabetes
AI-READI Flagship Dataset of Type 2 Diabetes
The AI-READI dataset consists of multimodal data collected from individuals with and without Type 2 Diabetes Mellitus (T2DM) and harmonized across three data collection sites. Designed for future AI/ML studies, recruitment aimed to achieve approximately equal distribution of participants across diabetes severity. A standardized data acquisition protocol spans multiple domains (surveys, physical measurements, clinical data, imaging, wearable devices, ECG, laboratory results, blood glucose, and environmental variables such as home air quality) to enable downstream AI/ML analyses that are not feasible with claims or electronic health records alone. Some non-sensitive data are publicly downloadable under a license agreement; the full dataset, including sensitive elements, is available under a data use agreement (controlled access). As enrollment is ongoing, pilot release(s) and periodic updates may not be balanced across groups.
en
  • AI-READI
  • Type 2 Diabetes Mellitus
  • T2DM
  • multimodal
  • harmonized
  • AI
  • machine learning
  • surveys
  • clinical measurements
  • retinal imaging
  • ECG
  • wearable devices
  • laboratory results
  • blood glucose
  • environmental data
  • FAIRhub
  • Response
    Enable downstream AI/ML analyses across multiple domains that are not feasible using existing data sources such as claims or electronic health records alone by providing harmonized, multi-site, multi-domain data.
  • Description
    • Controlled-access data include 5-digit ZIP code, sex, race, ethnicity, genetic sequencing data, past health records, medications, and traffic and accident reports.
  • Description
    • Past health records and related clinical information may include PHI and data protected by medical confidentiality; these are under controlled access.
Description
  • Public dataset excludes sensitive personal health data and is available under a license agreement; full dataset includes sensitive elements and is accessible only under a data use agreement (controlled access).
  1. Description
    • Data were collected directly from participants (surveys, physical and clinical measurements, wearable devices, imaging, and ECG), and some elements were obtained or derived from existing records (e.g., past health records) and genetic sequencing.
    Was Directly Observed
    yes
    Was Reported By Subjects
    yes
    Was Inferred Derived
    yes
  • Description
    • Standardized, harmonized acquisition protocol across three sites using clinical instruments (e.g., monofilament testing), imaging devices (retinal imaging), ECG, wearable devices, laboratory testing for blood and urine, participant surveys, and environmental sensing (home air quality).
  • Description
    • Documentation references a pilot study phase and ongoing enrollment; versions of the dataset and documentation are maintained per release.
  • Description
    • Domain-specific processing and metadata are described in the dataset documentation (e.g., file formats, standards, and example outputs); details vary by data domain.
  • External Resources
    • Dataset landing page on the FAIRhub data portal
    Restrictions
    • Access to public subset requires acceptance of a license agreement; full dataset requires execution of a data use agreement.
  • Description
    • AI-READI Project team (see dataset documentation and contact links)
DescriptionDistribution FormatsIDIs Data SplitIs SubpopulationName
Non-sensitive subset available for public download under a license agreement. Includes surveys, bloo...{'description': ['FAIRhub data portal public access (license-gated)']}ai-readi-public-subsetnonoAI-READI Public Dataset Subset
Full dataset available under a data use agreement. Includes sensitive elements such as 5-digit ZIP c...{'description': ['FAIRhub data portal controlled access (requires data use agreement)']}ai-readi-full-controllednonoAI-READI Full Dataset (Controlled Access)
Version associated with pilot study participants; documentation notes that pilot release(s) may not ...ai-readi-v1-0nonoAI-READI Dataset v1.0 (Pilot Phase)
A newer dataset version is indicated on the documentation site; details are provided in versioned do...ai-readi-v2-0-0nonoAI-READI Dataset v2.0.0
🚀

Uses

What (other) tasks could the dataset be used for?

  • Response
    Development and evaluation of AI/ML methods on multimodal T2DM-related data, including modeling, prediction, and analysis across clinical, imaging, wearable, laboratory, and environmental domains.
  • Description
    • Potential imbalance in pilot release(s) may affect fairness or generalizability; users should assess representation across groups and mitigate bias in downstream analyses.
Description
  • Public dataset: available for download upon agreement with a license that defines permitted uses.
  • Full dataset: accessible only by entering into a data use agreement (controlled access).
📤

Distribution

How will the dataset be distributed?

Description
  • Dataset and documentation are versioned; the site indicates multiple versions (e.g., v1.0 and v2.0.0) with per-version documentation accessible via a version selector.
🔄

Maintenance

How will the dataset be maintained?

2025-10-21
Description
  • Enrollment is ongoing; pilot data release with periodic updates to data releases is planned.
Generated on 2025-11-09 10:17:34 using Bridge2AI Data Sheets Schema