MedXAI: A Retrieval-Augmented and Self-Verifying Framework for Knowledge-Guided Medical Image Analysis

TL;DR

MedXAI is a framework that combines deep vision models with expert medical knowledge to enhance interpretability, robustness, and performance in medical image diagnosis, especially under domain shifts and rare conditions.

Contribution

It introduces a unified, knowledge-guided framework that improves generalization, reduces bias against rare classes, and provides human-understandable explanations without relying on post-hoc interpretability methods.

Findings

3% improvement in cross-domain generalization

10% increase in F1 score for rare classes

Consistent performance gains across multiple datasets

Abstract

Accurate and interpretable image-based diagnosis remains a fundamental challenge in medical AI, particularly under domain shifts and rare-class conditions. Deep learning models often struggle with real-world distribution changes, exhibit bias against infrequent pathologies, and lack the transparency required for deployment in safety-critical clinical environments. We introduce MedXAI (An Explainable Framework for Medical Imaging Classification), a unified expert knowledge based framework that integrates deep vision models with clinician-derived expert knowledge to improve generalization, reduce rare-class bias, and provide human-understandable explanations by localizing the relevant diagnostic features rather than relying on technical post-hoc methods (e.g., Saliency Maps, LIME). We evaluate MedXAI across heterogeneous modalities on two challenging tasks: (i) Seizure Onset Zone localization from resting-state fMRI, and (ii) Diabetic Retinopathy grading. Ex periments on ten multicenter datasets show consistent gains, including a 3% improvement in cross-domain generalization and a 10% improvmnet in F1 score of rare class, substantially outperforming strong deep learning baselines. Ablations confirm that the symbolic components act as effective clinical priors and regularizers, improving robustness under distribution shift. MedXAI delivers clinically aligned explanations while achieving superior in-domain and cross-domain performance, particularly for rare diseases in multimodal medical AI.