Echo2ECG: Enhancing ECG Representations with Cardiac Morphology from Multi-View Echos
Researchers have developed Echo2ECG, a novel multimodal self-supervised learning framework designed to enhance electrocardiography (ECG) representations by incorporating cardiac morphology data from multi-view echocardiography (Echo). This approach aims to enable the prediction of morphological phenotypes, such as Left ventricular ejection fraction, directly from ECGs, which is not possible with traditional ECG analysis alone. The framework was evaluated on its ability to classify cardiac phenotypes and retrieve similar Echo studies using ECG queries, demonstrating superior performance over existing unimodal and multimodal baselines. AI
IMPACT This research could lead to more accessible and earlier health screenings by enabling the prediction of cardiac morphology from standard ECGs.