AI-enabled imaging, ECG, and EHR biomarkers for earlier diagnosis and sharper risk assessment.
Section of Cardiovascular Medicine · Yale School of Medicine
We develop and test AI-enabled digital biomarkers for cardiovascular disease, using computer vision and statistical machine learning to read signal from imaging, ECG, and clinical data. The goal is practical: earlier diagnosis, better risk stratification, and tools that can be used in routine care.
Building imaging-derived features from echocardiography, CT, and ECG that capture disease biology not visible on routine interpretation, then testing whether they improve clinical decisions.
Studying how epicardial and perivascular fat carry early cardiovascular risk through radiomic and deep learning analysis of routine CT scans.
Using data-driven subgroup discovery to identify who benefits most from therapy and to improve how cardiovascular trials are designed, enriched, and interpreted.
Linking imaging, electrophysiology, and EHR data so AI tools can be evaluated where they will actually be used: at the point of care.
Undergraduate and graduate students, medical students, residents, fellows, postdoctoral researchers, and faculty collaborators are all welcome to reach out, especially if you want to build AI tools grounded in real clinical questions.