MReg: A Novel Regression Model with MoE-based Video Feature Mining for Mitral Regurgitation Diagnosis

2025-06-30

AI Summary

Current intelligent diagnostic methods for mitral regurgitation (MR) often misalign with clinical workflows, limiting both accuracy and interpretability. To address this, we propose a regression-based automatic MR grading framework specifically designed for four-chamber color Doppler echocardiographic video sequences. Instead of treating MR severity as discrete classes, our method models it as a continuous variable—better reflecting its inherently graded clinical nature. We introduce a feature selection and magnification mechanism that emulates physician visual assessment logic, alongside a Mixture-of-Experts (MoE)-inspired feature aggregation module and attention mechanisms to enhance class-level representation learning and precise regurgitant jet localization. Evaluated on a large in-house dataset of 1,868 cases, our approach significantly outperforms existing weakly supervised and fully supervised classification baselines, achieving superior MR severity estimation accuracy and improved decision interpretability.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Computer Vision: Multi-modal Vision
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Color Doppler echocardiography is a crucial tool for diagnosing mitral regurgitation (MR). Recent studies have explored intelligent methods for MR diagnosis to minimize user dependence and improve accuracy. However, these approaches often fail to align with clinical workflow and may lead to suboptimal accuracy and interpretability. In this study, we introduce an automated MR diagnosis model (MReg) developed on the 4-chamber cardiac color Doppler echocardiography video (A4C-CDV). It follows comprehensive feature mining strategies to detect MR and assess its severity, considering clinical realities. Our contribution is threefold. First, we formulate the MR diagnosis as a regression task to capture the continuity and ordinal relationships between categories. Second, we design a feature selection and amplification mechanism to imitate the sonographer's diagnostic logic for accurate MR grading. Third, inspired by the Mixture-of-Experts concept, we introduce a feature summary module to extract the category-level features, enhancing the representational capacity for more accurate grading. We trained and evaluated our proposed MReg on a large in-house A4C-CDV dataset comprising 1868 cases with three graded regurgitation labels. Compared to other weakly supervised video anomaly detection and supervised classification methods, MReg demonstrated superior performance in MR diagnosis. Our code is available at: https://github.com/cskdstz/MReg.

Problem

• Automated MR diagnosis model for mitral regurgitation severity assessment
• Feature mining to align with clinical workflow and accuracy
• Regression task to capture ordinal relationships in MR categories

Innovation

• Regression model for MR diagnosis continuity
• Feature selection mimicking sonographer logic
• Mixture-of-Experts for category-level feature extraction