Cross-sectional design and protocol for Artificial Intelligence Ready and Equitable Atlas for Diabetes Insights...

Area of Science:

• Medical Informatics
• Public Health
• Artificial Intelligence in Medicine

Background:

• Type 2 diabetes mellitus (T2DM) affects 10.5% of the global adult population, contributing significantly to adverse health outcomes.
• Promoting health resilience and salutogenesis in T2DM is crucial due to its high prevalence and associated health risks.

Purpose of the Study:

• To establish standards, best practices, and guidelines for collecting, preparing, and sharing data for AI/ML applications in T2DM research.
• To facilitate the use of AI/ML for studying salutogenesis, the transition from T2DM to health resilience.
• To serve as a model for developing future medical/health datasets for AI/ML.

Main Methods:

• A cross-sectional study targeting 4000 participants aged 40 and older, balanced by race/ethnicity, T2DM status, and sex.
• Multivariable data collection across over 10 domains, including clinical measurements, imaging, cognitive function, continuous glucose monitoring, physical activity, environmental factors, and psychosocial variables.
• Data collection conducted at three study sites: Birmingham, Alabama; San Diego, California; and Seattle, Washington.

Main Results:

• N/A - This abstract describes a data collection project, not specific findings from data analysis.

Conclusions:

• AI-READI provides a robust, equitable dataset to advance AI/ML in understanding T2DM salutogenesis.
• The project disseminates principles for creating AI/ML-ready health datasets, promoting FAIR data principles.
• Encourages researchers to utilize the AI-READI dataset to generate novel insights and publish findings.