Multimodal Stepwise Clinically-Guided Attention Learning for Pathological Complete Response Prediction in Breast Cancer

2026-05-08

AI Summary

This study addresses the challenge of predicting pathological complete response (pCR) to neoadjuvant therapy in breast cancer, which is hindered by severe class imbalance and limited cross-center generalizability. The authors propose a multimodal, two-stage, clinically guided attention learning framework that emulates clinical reasoning: it first extracts global imaging features, then leverages anatomically consistent tumor regions to guide spatial attention toward diagnostically relevant lesions, and finally integrates multiparametric MRI with clinical variables for decision-making. By embedding medical prior knowledge into the attention mechanism, the method significantly enhances sensitivity to the minority class and improves cross-center generalization while producing anatomically coherent, interpretable attention maps. The approach outperforms non-guided single-stage baseline models, maintaining high specificity without compromising predictive performance.

Abstract

Pathological complete response (pCR) is a key prognostic factor in breast cancer patients undergoing neoadjuvant therapy, strongly associated with long-term survival and treatment personalization. However, accurate pre-treatment pCR prediction remains challenging due to severe class imbalance and limited generalizability across diverse clinical settings. In this work, we propose a multimodal stepwise clinically-guided attention learning framework for pCR prediction from breast magnetic resonance imaging (MRI), designed to address these limitations through medically grounded spatial guidance and multimodal integration. The approach follows a stepwise training strategy inspired by physician reasoning: the model first learns global discriminative imaging patterns, then attention mechanisms are introduced to constrain the network toward tumor regions, and finally clinical variables are integrated to refine decision-making. This guidance strategy encourages prioritization of task-relevant features, improving identification of responders despite their limited representation in the dataset. Moreover, grounding attention in anatomically consistent tumor regions reduces reliance on dataset-specific patterns, thereby enhancing cross-institutional generalization. The framework is evaluated through external validation across heterogeneous MRI cohorts. Compared to non-guided single-stage baselines, the proposed approach improves sensitivity while maintaining competitive specificity, and produces anatomically coherent attention maps that support interpretation of the model's predictions. These findings highlight the potential of clinically-guided multimodal attention learning for robust and generalizable pCR prediction in breast cancer.

Problem

• pathological complete response
• breast cancer
• class imbalance
• generalizability
• neoadjuvant therapy

Innovation

• clinically-guided attention
• multimodal learning
• stepwise training
• pathological complete response
• cross-institutional generalization