Paper Overview
Field: Computer Vision Authors: Guang Yang, Wentian Xu, Siyu Wang, Betty Raman, Lei Li, Vicente Grau Published: 2026-07-16 arXiv: 2607.15268
Summary
Myocardial infarction (MI) remains a leading cause of mortality worldwide. Echocardiography (Echo) is a widely available modality for MI assessment, where regional wall motion abnormality is a key indicator. Prior learning-based methods for myocardial motion analysis often use handcrafted descriptors or densely supervised estimation, but the need for extensive annotation limits applicability.
Foundation models have recently improved vision-based Echo analysis; however, most methods operate on single views, and segment-level localization remains unreliable under view-dependent ambiguity, especially in apical views.
To address this, the authors propose MCF-Net, a novel motion-guided multi-view fusion framework that fuses myocardial motion cues with foundation model representations to localize infarction:
- Visual features are extracted using EchoPrime, a pre-trained Echo foundation model shared across dual views.
- Cardiac motion is modeled with extremely sparse supervision: a single annotated template frame is propagated across videos to initialize point tracking, avoiding dense annotation.
- Motion-derived segment-aware soft masks provide coarse spatial priors that selectively enhance features of challenging myocardial segments.
- A motion-conditioned fusion mechanism integrates motion and vision across views, refining predictions without overriding strong appearance cues.
Results
On segment-level MI localization, MCF-Net achieves 72.4% F1 and 84.9% accuracy, outperforming state-of-the-art motion-only, vision-only, and fusion baselines.
--- *Auto-collected on 2026-07-18*