What is the clinical evidence from this study?

Study design: Observational. Population: Type 2 diabetes mellitus (n=2280). Intervention: Multimodal data collection. Primary outcome: Dataset generation.

What does this research mean for the field?

The AI-READI project provides a large, standardized, multimodal, and AI-ready dataset of individuals across the type 2 diabetes spectrum to facilitate machine learning discoveries. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

The aim is to create a high-quality multimodal dataset for studying type 2 diabetes mellitus.

June 7, 2026

2301-P: The Artificial Intelligence Ready and Exploratory Atlas for Diabetes Insights (AI-READI) Dataset: Purpose, Design, Availability, and Use

Key Result

The AI-READI project has released a standardized, multimodal dataset from 2,280 participants aged ≥ 40 years to support artificial intelligence research in type 2 diabetes mellitus.

Key Points

The aim is to create a high-quality multimodal dataset for studying type 2 diabetes mellitus.
Recruited 4000 persons aged ≥ 40 years with various categories of type 2 diabetes.
Data is collected from sites in Birmingham, San Diego, and Seattle.
Included blood derivatives and multiple health domains relevant to T2DM.
2280 participants have data released optimized for AI/ML research.
Current data release contains 358,999 files, totaling 3.87 TB.
Final dataset release from ~4000 participants is expected by November 2026.

Study Design

Type

Observational (n=2,280)

Multicenter

Yes

Structured PICO

Population

An observational dataset collection of 2,280 participants aged ≥ 40 years, balanced across four categories of T2DM severity, to support AI discoveries.

The AI-READI project provides a large, multimodal, AI-ready dataset of individuals with and without type 2 diabetes to support future machine learning discoveries.

Abstract

Introduction and Objective: The ability to understand and affect the course of complex, multi-system diseases like diabetes has been limited by a lack of well-designed, high-quality and large multimodal datasets. The NIH Bridge2AI AI-READI project (aireadi.org) aims to address this shortfall by generating an AI-ready dataset to support AI discoveries in type 2 diabetes mellitus (T2DM). Methods: Our goal is to recruit 4000 persons ≥ 40 years of age balanced for 4 categories of T2DM: non-diabetic, prediabetes/lifestyle-controlled, controlled by oral-medications/non-insulin injectables and insulin-dependent. Data collection sites are located in Birmingham AL, San Diego CA, and Seattle WA. The variable domains of the dataset encompass many biomedical and behavioral aspects of health often impacted in T2DM (Figure 1). Blood derivatives, including serum, plasma, RNA and Peripheral Blood Mononuclear Cells, are available to researchers for future research. Results: As of November 2025, data from 2280 participants has been released that is standardized and optimized for AI/ML research. The current release includes 358,999 files (3.87 TB) of data and is available to researchers through a registered public access or a controlled access database, depending on the variables requested. A final dataset release from ~4000 participants is scheduled by November, 2026. Conclusion: This flagship dataset may enable new AI discoveries in T2DM. Disclosure D. Matthies: None. J. Owen: None. G. McGwin: None. C. Owsley: Consultant; Current; Johnson Current; Boehringer Ingelheim International GmbH, Roche Pharmaceuticals. Consultant; Current; Sanofi. S. Baxter: Consultant; Ended; Topcon Healthcare. L.M. Zangwill: Research Support; Current; Heidelberg Engineering, Carl Zeiss Meditec, Topcon Medical Systems, Optovue Inc, iCare Inc, Optomed, EssilorLuxotica. Other - Travel; Current; AISight Health. C. Lee: None. A. Lee: Research Support; Current; Regeneron Pharmaceuticals Inc. Research Support; Ended; Topcon Healthcare. Consultant; Current; Sanofi-Aventis U.S. Consultant; Ended; Boehringer Ingelheim International GmbH. Consultant; Current; Johnson Ended; Zeiss. Consultant; Current; Gyroscope. Research Support; Current; Santen. Funding National Institutes of Health (OT2OD032644)

Mark Helpful

Bookmark

Relay