Skip to contentSkip to filters

We use cookies for analytics.

Seminar, Hosted by Stanford MedAI

PanEcho: Toward Complete AI-Enabled Echocardiography Interpretation

Gregory Holste

University of Texas, Austin

Recording

1 h

Abstract

Echocardiography is the most widely used cardiac imaging modality, but it relies on expert manual interpretation of complex ultrasound videos, a scarce resource given the shortage of cardiologists. Artificial intelligence could automate echocardiogram interpretation, widening access to cardiovascular care and freeing cardiologists' time for patients. This talk traces the evolution of AI-enabled echocardiography from specialized single-task models to comprehensive automated systems, beginning with self-supervised learning and aortic stenosis detection from a single view of the heart and culminating in PanEcho, an end-to-end system for multi-view echocardiogram interpretation that mirrors a real clinical workflow. It closes with what is needed to translate these advances into clinical practice and to democratize access to cardiovascular care worldwide.

Topics