Article:Moon-Seung Soh
Deep Learning Detects LV Diastolic Dysfunction from Standard ECGs
A research team led by Moon-Seung Soh developed a vision transformer model to identify left ventricular diastolic dysfunction (LVDD) from standard 12-lead ECGs. With a high negative predictive value, the model shows potential as a scalable, non-invasive initial screening tool.
From Video: Weekly AI Updates: Local Models, GPT-6 Astra, and Agent Architectures