Introduction to AI-READI, Studying Salutogenesis in T2DM (dkNET Presentation)

The Abstract of the talk is provided below. More information is available at https://dknet.org/about/webinar. Abstract The AI-READI (Artificial Intelligence Ready and Equitable Atlas for Diabetes Insights, https://aireadi.org) project, funded by the NIH Common Fund’s Bridge2AI Program, aims to develop a multimodal dataset specifically designed to be AI-ready for the study of salutogenesis in Type 2 Diabetes Mellitus (T2DM). Despite advancements in diabetes care, limited knowledge exists on how individuals with T2DM may revert to health. AI-READI team is building this dataset from a diverse cohort of 4,000 participants, ensuring it is structured for immediate use in machine learning algorithm training and analysis. The project emphasizes ethical and equitable data collection, adherence to FAIR (Findable, Accessible, Interoperable, Reusable) data principles, and establishing best practices for data sharing and management. By focusing on AI-readiness, the dataset will enable rapid application of machine learning to uncover novel insights into effective treatment strategies. This presentation will introduce the AI-READI project, present the dataset, demonstrate how to request the datasets and explore potential research questions that can be addressed using machine learning, such as identifying predictors of health improvement in T2DM, understanding disease progression, and investigating the impact of various risk factors. The top 3 key questions that Bridge2AI AI-READI datasets can answer: How can we better understand Type 2 Diabetes (T2DM) heterogeneity? What are the connections between multi-organ function in T2DM, and how are the kidney, heart, eye, brain interlinked? How do interactions between environmental factors (e.g. air pollution) drive outcomes in T2DM?