Multimodal Survival Modeling and Fairness-Aware Clinical Machine Learning for 5-Year Breast Cancer Risk Prediction

2026-02-25

AI Summary

This study addresses common challenges in clinical risk prediction models—such as poor calibration, limited generalizability, and subgroup bias—when applied to high-dimensional multimodal breast cancer data. The authors propose a governance-oriented, reproducible multimodal survival modeling framework that integrates clinical variables, transcriptomic profiles, and copy number variation features. The pipeline incorporates variance- and sparsity-based filtering, dimensionality reduction, and survival modeling via CoxNet (an elastic net–regularized Cox model) and XGBoost gradient-boosted survival trees, with rigorous hyperparameter optimization. Emphasizing calibration, fairness, robustness, and reproducibility, the framework achieves time-dependent AUCs of 96.6% and 92.5% for five-year survival prediction. Fairness audits further demonstrate stable performance across diverse clinical subgroups.

Technology Category

• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision
• Data Mining & Knowledge Management: Mining of Visual, Multimedia & Multimodal Data

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Social Networks and Social Media: Fairness and bias in social network and social media analysis
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Clinical risk prediction models often underperform in real-world settings due to poor calibration, limited transportability, and subgroup disparities. These challenges are amplified in high-dimensional multimodal cancer datasets characterized by complex feature interactions and a p >> n structure. We present a fully reproducible multimodal machine learning framework for 5-year overall survival prediction in breast cancer, integrating clinical variables with high-dimensional transcriptomic and copy-number alteration (CNA) features from the METABRIC cohort. After variance- and sparsity-based filtering and dimensionality reduction, models were trained using stratified train/validation/test splits with validation-based hyperparameter tuning. Two survival approaches were compared: an elastic-net regularized Cox model (CoxNet) and a gradient-boosted survival tree model implemented using XGBoost. CoxNet provides embedded feature selection and stable estimation, whereas XGBoost captures nonlinear effects and higher-order interactions. Performance was assessed using time-dependent area under the ROC curve (AUC), average precision (AP), calibration curves, Brier score, and bootstrapped 95 percent confidence intervals. CoxNet achieved validation and test AUCs of 98.3 and 96.6, with AP values of 90.1 and 80.4. XGBoost achieved validation and test AUCs of 98.6 and 92.5, with AP values of 92.5 and 79.9. Fairness diagnostics showed stable discrimination across age groups, estrogen receptor status, molecular subtypes, and menopausal state. This work introduces a governance-oriented multimodal survival framework emphasizing calibration, fairness auditing, robustness, and reproducibility for high-dimensional clinical machine learning.

Problem

• clinical risk prediction
• multimodal data
• subgroup disparities
• high-dimensional data
• model calibration

Innovation

• multimodal survival modeling
• fairness-aware machine learning
• high-dimensional clinical data
• CoxNet
• XGBoost survival