A deep learning pipeline for PAM50 subtype classification using histopathology images and multi-objective patch selection

Revolutionizing Breast Cancer Diagnostics with AI

Breast cancer is a highly heterogeneous disease with diverse molecular profiles. Recent AI models have attempted to predict PAM50 subtype, as a standard for classifying breast cancer into intrinsic subtypes, from histopathology images, most depend on random patch sampling that introduces redundancy and restricts the model performance. In this study, we introduce a novel optimization-driven deep learning framework that aims to reduce reliance on costly molecular assays by directly predicting PAM50 subtypes from H&E-stained whole-slide images (WSIs). Our method jointly optimizes patch informativeness, spatial diversity, uncertainty, and patch count by combining the non-dominated sorting genetic algorithm II (NSGA-II) with Monte Carlo dropout-based uncertainty estimation. The proposed method can identify a small but highly informative patch subset for classification. We used a ResNet18 backbone for feature extraction and a fully connected head for classification. For evaluation, we used the internal TCGA-BRCA dataset as the training cohort and the external CPTAC-BRCA dataset as the test cohort. On the internal dataset, an F1-score of 0.8964 and an AUC of 0.9865 using 627 WSIs from the TCGA-BRCA cohort were achieved. The performance of the proposed approach on the external validation dataset showed an F1-score of 0.7995 and an AUC of 0.9523. These findings indicate that the proposed optimization-guided, uncertainty-aware patch selection can achieve high performance and improve the computational efficiency of histopathology-based PAM50 classification compared to existing methods, suggesting a scalable imaging-based replacement that has the potential to support clinical decision-making.