Explainability in Generative Medical Diffusion Models: A Faithfulness-Based Analysis on MRI Synthesis
This work addresses the problem of interpretability for healthcare practitioners using generative AI, though it is incremental as it applies existing explainability methods to a new domain.
The study tackled the lack of transparency in generative diffusion models for medical imaging by developing a faithfulness-based explainability framework to analyze MRI synthesis, finding that Enhanced ProtoPNet achieved the highest faithfulness score of 0.1534 for reliable insights into the generative process.
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.