Researchers are releasing the flagship dataset from a study on biomarkers and environmental factors that may influence the development of type 2 diabetes. The study includes participants with various stages of the condition, revealing a tapestry of information distinct from previous research. Data collected include environmental sensor readings, survey responses, depression scales, eye-imaging scans, and traditional biological measures, all intended to be analyzed by artificial intelligence for novel insights on risks, preventive measures, and disease pathways.
Dr. Cecilia Lee, a professor at the University of Washington School of Medicine, highlights the heterogeneity among type 2 diabetes patients, indicating that not everyone is dealing with the same issues. The collected data from 1,067 participants represent just 25% of the study’s total expected enrollees. The AI-READI initiative aims to collect and share AI-ready data for global scientists to analyze for new insights into health and disease. The data were released in a paper in the journal Nature Metabolism and aim to be more racially and ethnically diverse than previous studies.
Dr. Aaron Lee, the project’s principal investigator, notes the collaborative effort of seven institutions and multidisciplinary teams working together to collect unbiased data and ensure its security. The study sites in Seattle, San Diego, and Birmingham are enrolling 4,000 participants with balanced criteria for race/ethnicity, disease severity, and sex. The aim is to study not only disease pathogenesis and risk factors but also factors that contribute to health, leading to novel discoveries about type 2 diabetes.
By collecting in-depth data from a large number of people, the researchers hope to create pseudo health histories to understand how individuals progress from disease to health and vice versa. The data, hosted on a custom online platform, are available in two sets – a controlled-access set with usage agreements and a publicly available version stripped of HIPAA-protected information. The pilot data release has been downloaded by over 110 research organizations worldwide, with researchers required to verify their identity and agree to ethical usage terms.
The AI-READI Consortium includes various institutions such as the University of Washington School of Medicine, University of Alabama at Birmingham, University of California San Diego, and Johns Hopkins University. The project, based at UW Medicine in Seattle, was supported by NIH grants and aims to provide valuable insights into type 2 diabetes through comprehensive data analysis. The researchers hope that by studying both disease and health factors, they can uncover new information that may lead to improved prevention and treatment strategies for diabetes.