Federated Deep Learning for Privacy-Preserving Cardiovascular Disease Risk Prediction

2026-07-09

AI Summary

This study addresses the challenge of cross-institutional cardiovascular disease risk prediction, which is hindered by data privacy regulations and heterogeneity that preclude sharing of individual-level data. The work presents the first implementation of federated deep survival analysis on real-world heterogeneous medical cohorts—the Lifelines and Rotterdam Study—enabling collaborative model training without exchanging raw data. By preserving data privacy while maintaining model performance, the proposed approach achieves improved predictive accuracy, increasing the C-statistic from 0.728 to 0.739 in the Rotterdam Study and from 0.783 to 0.787 in Lifelines. These results demonstrate the effectiveness and added value of federated learning in multi-center clinical settings.

Abstract

Cardiovascular disease risk prediction models often rely on data from a single institution or centrally pooled datasets. Extending these models across institutions could be limited by privacy regulations and constraints on sharing patient-level data. Federated learning enables collaborative model development without transferring sensitive patient data, but its application in healthcare remains challenging because datasets often differ in size, population characteristics, and outcome definitions. In this study, we present a federated deep learning approach for privacy-preserving cardiovascular disease risk prediction that integrates two population-based cohorts with different characteristics: Lifelines, including 148,230 participants meeting the study inclusion criteria with self-reported outcomes, and the Rotterdam Study, including a smaller cohort of 10,155 participants with digitally linked clinical outcomes. Model performance was primarily evaluated on the Rotterdam Study because of its complete follow-up. Deep survival models trained using federated learning achieved higher predictive performance than models trained locally without federation. For the Rotterdam Study, the C-statistic increased from 0.728 (95% CI: 0.717-0.739) to 0.739 (95% CI: 0.728-0.749). For Lifelines, the C-statistic increased from 0.783 (95% CI: 0.775-0.791) to 0.787 (95% CI: 0.780-0.792). These findings suggest that federated deep learning across heterogeneous cohorts can improve cardiovascular disease risk prediction while preserving the privacy of individual-level patient data.

Problem

• Federated Learning
• Privacy-Preserving
• Cardiovascular Disease Risk Prediction
• Heterogeneous Cohorts
• Deep Learning

Innovation

• federated learning
• deep survival model
• cardiovascular disease risk prediction
• privacy-preserving AI
• heterogeneous cohorts