Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces BVFLMSP, a Bayesian Vertical Federated Learning framework for multimodal survival analysis that addresses privacy concerns and provides uncertainty estimates. The framework employs a Split Neural Network architecture where clients model individual data modalities using Bayesian neural networks, and a central server aggregates representations for survival risk prediction. By integrating differential privacy mechanisms, BVFLMSP achieves formal privacy guarantees while demonstrating improved discrimination performance (up to 0.02 higher C-index) compared to centralized multimodal baselines across various multimodal settings.
You can now train multimodal survival models across federated datasets with formal privacy guarantees, all while getting uncertainty estimates that improve prediction accuracy.
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.