BVFLMSP : Bayesian Vertical Federated Learning for Multimodal Survival with Privacy

TL;DR

BVFLMSP is a novel Bayesian federated learning framework for multimodal survival analysis that enhances privacy and provides uncertainty estimates, outperforming existing methods in predictive accuracy.

Contribution

It introduces a privacy-preserving Bayesian VFL approach with a split neural network architecture for multimodal survival prediction, improving performance and robustness.

Findings

Outperforms single modality and centralized baselines in C-index by up to 0.02.

Effectively balances privacy and predictive performance under various privacy budgets.

Provides uncertainty estimates alongside survival predictions.

Abstract

Multimodal time-to-event prediction often requires integrating sensitive data distributed across multiple parties, making centralized model training impractical due to privacy constraints. At the same time, most existing multimodal survival models produce single deterministic predictions without indicating how confident the model is in its estimates, which can limit their reliability in real-world decision making. To address these challenges, we propose BVFLMSP, a Bayesian Vertical Federated Learning (VFL) framework for multimodal time-to-event analysis based on a Split Neural Network architecture. In BVFLMSP, each client independently models a specific data modality using a Bayesian neural network, while a central server aggregates intermediate representations to perform survival risk prediction. To enhance privacy, we integrate differential privacy mechanisms by perturbing client side representations before transmission, providing formal privacy guarantees against information leakage during federated training. We first evaluate our Bayesian multimodal survival model against widely used single modality survival baselines and the centralized multimodal baseline MultiSurv. Across multimodal settings, the proposed method shows consistent improvements in discrimination performance, with up to 0.02 higher C-index compared to MultiSurv. We then compare federated and centralized learning under varying privacy budgets across different modality combinations, highlighting the tradeoff between predictive performance and privacy. Experimental results show that BVFLMSP effectively includes multimodal data, improves survival prediction over existing baselines, and remains robust under strict privacy constraints while providing uncertainty estimates.