PANDA stands for Precision Analytics for Nutrition and Dietary Analysis. PANDA has 2 parts: AIDA - Artificial Intelligence for Dietary Assessment and DROID - Diet Recommendation Optimization to Impede Diabetes. Both parts will use continuous glucose monitors (CGMs) to accomplish their goal. The goal of AIDA is to use CGMs and artificial intelligence to determine what someone ate. Right now, we have to ask people what they ate, which requires people to remember what and how much they ate and takes over an hour! AIDA will solve this problem. The goal of DROID is to use CGMs and artificial intelligence to determine what diet pattern would be best for someone's blood glucose. This may be different from what they are eating right now and different from someone else's ideal dietary pattern. The PANDA study is pending funding.

GAIL stands for Gamified, Artificial-Intelligence-enabled Lifestyle change. The goal of the GAIL project is to develop a smartphone app that helps people improve their eating and physical activity habits through supportive, personalized conversations with an artificial intelligence (AI) coach. The app uses advanced language technology trained on peer-reviewed nutrition and exercise science to provide accurate, evidence-based guidance, while also being designed to communicate with empathy and motivational interviewing techniques that encourage lasting behavior change. GAIL includes game-like features to make healthy lifestyle changes engaging and rewarding. Importantly, the app is being built in partnership with community members, researchers, and clinicians, using feedback from focus groups to ensure it is practical, respectful, and responsive to the needs of the people it is meant to serve. Funded by the National Institutes of Health through The Ohio State University Clinical and Translational Sciences Institute.

This study examines how challenging social and economic conditions—often called adverse social determinants of health (SDoH), such as limited income, housing instability, or reduced access to care—affect outcomes for people hospitalized with heart failure. Using data from approximately 120,000 patients admitted with heart failure, we are developing computer-based risk prediction tools that incorporate both medical information and social risk factors to better estimate a patient’s risk of dying during their hospital stay. We are also studying whether patients facing more adverse social conditions tend to arrive at the hospital with more severe heart failure. Ultimately, our goal is to determine how much these social factors contribute to mortality risk, so that health systems can design more equitable care strategies and better support patients who face greater social disadvantage. This study is funded by the American Heart Association and uses their Get With The Guidelines data.

This study examines how preventive services delivered by Registered Dietitians—such as medical nutrition therapy and diabetes self-management education and support—affect the risk of diabetes-related complications, death, and healthcare costs. Using healthcare claims data from approximately 10 million people with diabetes, we are evaluating whether patients who receive these evidence-based preventive therapies experience better outcomes and lower spending over time. We are also identifying which modifiable risk factors—such as obesity, adverse social conditions, and structural inequities—contribute most strongly to complications and mortality risk. By understanding how each patient’s unique combination of risk factors shapes their individual risk profile, our goal is to help prioritize prevention strategies in a more personalized and equitable way, ensuring that resources are directed where they can have the greatest impact. This project uses data from the Komodo Healthcare Map and is funded by the Robert Wood Johnson Foundation.