Multimodal Attention-Aware Fusion for Diagnosing Distal Myopathy: Evaluating Model Interpretability and Clinician Trust

TL;DR

This paper introduces a multimodal attention-aware fusion model for diagnosing distal myopathy, improving accuracy and interpretability through an attention gate mechanism, validated by expert radiologists and benchmark datasets.

Contribution

It presents a novel fusion architecture combining global and local features with attention mechanisms, enhancing diagnostic performance and interpretability in medical imaging.

Findings

High classification accuracy on benchmark and proprietary datasets

Improved interpretability with saliency maps and attention mechanisms

Identified gaps in anatomical specificity and clinical usefulness

Abstract

Distal myopathy represents a genetically heterogeneous group of skeletal muscle disorders with broad clinical manifestations, posing diagnostic challenges in radiology. To address this, we propose a novel multimodal attention-aware fusion architecture that combines features extracted from two distinct deep learning models, one capturing global contextual information and the other focusing on local details, representing complementary aspects of the input data. Uniquely, our approach integrates these features through an attention gate mechanism, enhancing both predictive performance and interpretability. Our method achieves a high classification accuracy on the BUSI benchmark and a proprietary distal myopathy dataset, while also generating clinically relevant saliency maps that support transparent decision-making in medical diagnosis. We rigorously evaluated interpretability through (1) functionally grounded metrics, coherence scoring against reference masks and incremental deletion analysis, and (2) application-grounded validation with seven expert radiologists. While our fusion strategy boosts predictive performance relative to single-stream and alternative fusion strategies, both quantitative and qualitative evaluations reveal persistent gaps in anatomical specificity and clinical usefulness of the interpretability. These findings highlight the need for richer, context-aware interpretability methods and human-in-the-loop feedback to meet clinicians' expectations in real-world diagnostic settings.