Explanation-Aware Learning for Enhanced Interpretability in Biomedical Imaging

TL;DR

This paper introduces a method to incorporate explanation supervision into training deep neural networks for biomedical imaging, improving interpretability without sacrificing accuracy.

Contribution

It systematically analyzes how different explanation loss designs affect model interpretability and performance in medical image diagnosis.

Findings

Explanation supervision improves alignment of model explanations with clinical regions.

There is a trade-off between explanation quality and the strength of explanation loss.

The framework maintains predictive accuracy while enhancing interpretability.

Abstract

Deep neural networks for medical image diagnosis often achieve high predictive accuracy while relying on spurious or clinically irrelevant visual cues, limiting their trustworthiness in practice. Post-hoc explanation methods are widely used to visualize model decisions in the form of saliency maps; however, these explanations do not influence how models learn during training, allowing non-causal or confounding features to persist. This motivates the incorporation of explanation supervision directly into the training objective to guide model attention toward clinically meaningful regions and promote clinically grounded decision-making. This paper presents a systematic approach to integrate explanation loss into model training and analyzes how different explanation loss designs and supervision strengths influence both predictive performance and spatial faithfulness of explanations. To quantitatively assess interpretability, two complementary explanation performance metrics-annotation coverage and saliency precision-are introduced, enabling rigorous evaluation beyond qualitative visualization. Our experimental results reveal a clear trade-off between explanation quality and explanation loss coefficients. Furthermore, quantitative statistical analysis yields consistently improved explanation alignment while maintaining comparable accuracy. Experiments were conducted on annotated chest X-ray datasets; however, the proposed framework is applicable to a broad range of annotated biomedical imaging modalities. Overall, these findings demonstrate that explanation supervision is not a monolithic design choice and provide practical guidance for incorporating explanation loss into training objectives under noisy clinical annotations.