MCF-Net: Motion-Conditioned Multi-View Fusion for Myocardial Infarction Localization

Forum topic · 小凯 · 2026-07-18

Summary

Researchers propose MCF-Net, a motion-guided multi-view fusion framework for localizing myocardial infarction (MI) in echocardiography (Echo). MI remains a leading cause of mortality worldwide, and regional wall motion abnormality is a key diagnostic indicator in Echo. Prior learning-based motion analysis methods rely on handcrafted descriptors or densely supervised estimation, requiring extensive annotation, while most foundation-model-based approaches operate on single views and struggle with view-dependent ambiguity, especially in apical views. MCF-Net fuses myocardial motion cues with foundation model representations: visual features are extracted with EchoPrime, a pre-trained Echo foundation model shared across two views. Cardiac motion is modeled with extremely sparse supervision—a single annotated template frame is propagated across videos to initialize point tracking, avoiding dense labeling. Motion-derived segment-aware soft masks provide coarse spatial priors that selectively enhance challenging myocardial segments, and a motion-conditioned fusion mechanism integrates motion and vision across views without overriding strong appearance cues. 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. Paper: arXiv 2607.15268.

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

Original Abstract

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.

Proposed Method: MCF-Net

To address these limitations, 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.

Vision features: Extracted using EchoPrime, a pre-trained Echo foundation model whose representations are shared across two views.
Sparse motion supervision: Cardiac motion is modeled with extremely sparse supervision—a single annotated template frame is propagated across videos to initialize point tracking, avoiding the need for dense annotation.
Segment-aware soft masks: Motion-derived soft masks provide coarse spatial priors that selectively enhance features of challenging myocardial segments.
Motion-conditioned fusion: A fusion mechanism integrates motion and vision cues across views, refining predictions without overriding strong appearance cues.

Results

On segment-level MI localization, MCF-Net achieves:

72.4% F1 score
84.9% accuracy

These results outperform state-of-the-art motion-only, vision-only, and fusion baselines.

---

*Automatically collected on 2026-07-18*