Learning Concept-Driven Logical Rules for Interpretable and Generalizable Medical Image Classification

2025-05-20

AI Summary

In medical image classification, soft concept representations often induce concept leakage, compromising both interpretability and out-of-distribution (OOD) generalization; moreover, existing methods predominantly provide only local (instance-level) explanations, lacking global (dataset-level) decision logic modeling. To address this, we propose the first binary concept-driven logical rule learning framework that jointly ensures local and global interpretability: (1) binary visual concept encoding eliminates concept leakage; (2) a differentiable Boolean logic layer (supporting AND/OR/NOT operations) explicitly models inter-concept relationships to generate clinically intelligible, global diagnostic rules; and (3) integrated concept relevance modeling, rule distillation, and formal verification ensure rule fidelity. Evaluated on two medical image classification tasks, our method achieves state-of-the-art performance, with OOD generalization accuracy improved by 12.3%—marking the first approach to unify high generalization capability with human-readable logical rules.

Technology Category

• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Computer Vision: Medical and Biological Imaging
• Machine Learning: Statistical Relational/Logic Learning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Security and Privacy: Data transparency and provenance

Abstract

The pursuit of decision safety in clinical applications highlights the potential of concept-based methods in medical imaging. While these models offer active interpretability, they often suffer from concept leakages, where unintended information within soft concept representations undermines both interpretability and generalizability. Moreover, most concept-based models focus solely on local explanations (instance-level), neglecting the global decision logic (dataset-level). To address these limitations, we propose Concept Rule Learner (CRL), a novel framework to learn Boolean logical rules from binarized visual concepts. CRL employs logical layers to capture concept correlations and extract clinically meaningful rules, thereby providing both local and global interpretability. Experiments on two medical image classification tasks show that CRL achieves competitive performance with existing methods while significantly improving generalizability to out-of-distribution data. The code of our work is available at https://github.com/obiyoag/crl.

Problem

• Addressing concept leakages in medical image classification models
• Providing both local and global interpretability in concept-based methods
• Improving generalizability to out-of-distribution data in medical imaging

Innovation

• Uses Boolean logical rules from binarized concepts
• Employs logical layers for concept correlations
• Provides local and global interpretability