Explainability in Generative Medical Diffusion Models: A Faithfulness-Based Analysis on MRI Synthesis

TL;DR

This paper introduces a faithfulness-based explainability framework for generative diffusion models in MRI synthesis, enhancing transparency and trustworthiness in medical image generation.

Contribution

It proposes a novel faithfulness-based analysis linking prototype explainability methods to diffusion models, improving interpretability in medical imaging.

Findings

EPPNet achieves highest faithfulness score of 0.1534

Diffusion models can be made more transparent with faithfulness-based explanations

The framework provides reliable insights into the generative process

Abstract

This study investigates the explainability of generative diffusion models in the context of medical imaging, focusing on Magnetic resonance imaging (MRI) synthesis. Although diffusion models have shown strong performance in generating realistic medical images, their internal decision making process remains largely opaque. We present a faithfulness-based explainability framework that analyzes how prototype-based explainability methods like ProtoPNet (PPNet), Enhanced ProtoPNet (EPPNet), and ProtoPool can link the relationship between generated and training features. Our study focuses on understanding the reasoning behind image formation through denoising trajectory of diffusion model and subsequently prototype explainability with faithfulness analysis. Experimental analysis shows that EPPNet achieves the highest faithfulness (with score 0.1534), offering more reliable insights, and explainability into the generative process. The results highlight that diffusion models can be made more transparent and trustworthy through faithfulness-based explanations, contributing to safer and more interpretable applications of generative AI in healthcare.