GAFR-Net: A Graph Attention and Fuzzy-Rule Network for Interpretable Breast Cancer Image Classification

TL;DR

GAFR-Net addresses the challenge of interpretable breast cancer histopathology classification under limited annotations by integrating a similarity-based inter-sample graph with topology-aware descriptors and a differentiable fuzzy-rule reasoning module. The architecture combines multi-head graph attention for relational learning with explicit IF-THEN rules derived from topological features, enabling transparent decision paths. Empirical results on BreakHis, Mini-DDSM, and ICIAR2018 show state-of-the-art performance across magnifications while delivering interpretable reasoning, validated by ablation studies. This framework enhances generalization in weakly supervised medical image analysis and provides a practical, accountable tool for clinical decision support.

Abstract

Accurate classification of breast cancer histopathology images is pivotal for early oncological diagnosis and therapeutic intervention.However, conventional deep learning architectures often encounter performance degradation under limited annotations and suffer from a "blackbox" nature, hindering their clinical integration. To mitigate these limitations, we propose GAFRNet, a robust and interpretable Graph Attention and FuzzyRule Network specifically engineered for histopathology image classification with scarce supervision. GAFRNet constructs a similarity-driven graph representation to model intersample relationships and employs a multihead graph attention mechanism to capture complex relational features across heterogeneous tissue structures.Concurrently, a differentiable fuzzy-rule module encodes intrinsic topological descriptorsincluding node degree, clustering coefficient, and label consistencyinto explicit, human-understandable diagnostic logic. This design establishes transparent "IF-THEN" mappings that mimic the heuristic deduction process of medical experts, providing clear reasoning behind each prediction without relying on post-hoc attribution methods. Extensive evaluations on three benchmark datasets (BreakHis, Mini-DDSM, and ICIAR2018) demonstrate that GAFR-Net consistently outperforms various state-of-the-art methods across multiple magnifications and classification tasks. These results validate the superior generalization and practical utility of GAFR-Net as a reliable decision-support tool for weakly supervised medical image analysis.

Figures (6)

• Figure S1: Overall pipeline of the proposed GAFR-Net, consisting of graph construction, topological descriptor extraction, graph attention message passing, fuzzy-rule reasoning, and gating fusion for interpretable breast cancer histopathology classification.
• Figure S2: The intrinsic reasoning mechanism of GAFR-Net. (A) Input topological descriptors representing the local and global graph structure; (B) Feature mapping through trainable Gaussian membership functions to handle uncertainty; (C) Explicit "IF-THEN" logic activation that provides an accountable diagnostic path.
• Figure S3: Images of Breast Cancer Histopathology from the BreakHis Dataset at Four Magnifications
• Figure S4: Images of Breast Cancer Histopathology from the Mini-DDSM Dataset
• Figure S5: Images of Breast Cancer Histopathology from the ICIAR2018 Dataset