Cardiometabolic Imaging, Adipose Tissue, and Vascular Inflammation
We study how perivascular, epicardial, and systemic fat interacts with the cardiovascular system, and how it can be measured from routine imaging. Using CT radiomics and machine learning, we have established fat phenotyping as a scalable readout of cardiometabolic risk and early disease.
Digital Multimodality Biomarkers for Valvular Heart Disease
We build AI biomarkers for aortic stenosis that capture structural, myocardial, and hemodynamic remodeling from routine clinical data, enabling earlier detection of active disease and better timing of intervention.
Scalable AI-Enabled Screening for Underrecognized Cardiomyopathies
We build ECG and echocardiography tools to detect transthyretin amyloid and hypertrophic cardiomyopathy before overt presentation, designed for scalable use across health systems and real-world care.
Precision Clinical Trials and Digital Phenomapping
We learn signatures of individual treatment response, integrating EHR, imaging, and unstructured data to change how cardiovascular trials are run, designed, and interpreted.
Early Risk Stratification of Heart Failure and Cardiotoxicity
We apply AI to ECG and echocardiographic data to flag early myocardial injury, heart failure risk, and cardiotoxicity, supporting scalable surveillance and earlier prevention in high-risk patients.
Best practices in Cardiovascular AI
We develop frameworks and checklists for the rigorous development, evaluation, and deployment of cardiovascular AI, prioritizing transparency, equity, and clinical impact over model performance alone.
Featured Works
