Carl Yang, PhD
Assistant Professor,
Emory University
Featured in the Session: Rising Minds: AI Innovations in Diabetes—NIDDK Early-Career Investigator Symposium
When
Sunday, June 7
at 8:00 a.m. CT
Where
La Nouvelle Orleans C (Level 2)
Ernest N. Morial Convention Center
What is your presentation about?
This study bridges the gap between theoretical diabetes stratification and large-scale characterization of metabolic disease. We developed a joint learning framework that integrates future complication risk directly into the type 2 diabetes subtyping process. Leveraging large-scale electronic health record data, we applied the approach to 11,765 diverse participants from the National Institutes of Health (NIH) All of Us Research Program and validated our method in the UK Biobank, identifying five reproducible, outcome-driven subtypes. In addition to clinical trajectories, integration of whole-genome sequencing data revealed distinct polygenic architectures across groups, linking subtype-specific clinical patterns to coherent phenotype-genotype pathways. These results provide genetic support for the biological specificity of the subtypes and highlight mechanistic differences in type 2 diabetes progression.
How do you hope your presentation will impact diabetes research or care?
Our artificial intelligence (AI)-empowered analysis reveals critical phenotypes that traditional glucocentric classifications miss. We identified five reproducible subtypes that hierarchically organize into two broad meta-clusters. This stratification not only precisely recapitulates established phenotypes found in previous studies, but also reveals finer-grained, insightful heterogeneity.
How did you become involved with this area of diabetes research or care?
I was trained as a computer scientist focusing on AI/machine learning (ML) methodologies and I have a long family history of type 2 diabetes. I was fortunate to receive a K25 award from NIH National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) to support my in-depth application of AI/ML in diabetes research, especially around the topic of “Understanding Diabetes Heterogeneity via Mining Multimodality Interconnected Data.”