Uncertainty-Aware Foundation Models for Clinical Data
This work addresses the challenge of handling inherent uncertainty in clinical observations for healthcare foundation models, representing an incremental advancement by integrating uncertainty modeling into existing multimodal frameworks.
The paper tackled the problem of incomplete and heterogeneous clinical data by proposing an uncertainty-aware foundation model that represents patients as distributions over latent states, improving predictive performance, robustness under missing data, and uncertainty calibration across diverse clinical tasks.
Healthcare foundation models have largely followed paradigms from natural language processing and computer vision, emphasizing large scale pretraining and deterministic representations over heterogeneous clinical data. However, clinical observations are inherently incomplete, reflecting sparse, irregular, and modality dependent measurements of an underlying physiologic state. In this work, we propose a framework for uncertainty aware foundation modeling that represents each patient not as a point embedding, but as a distribution over plausible latent states. By learning set valued representations and enforcing consistency across partial views of the same patient, the model captures what is invariantly inferable while explicitly encoding epistemic uncertainty. We integrate this formulation with multimodal encoders and scalable self supervised objectives, combining reconstruction, contrastive alignment, and distributional regularization. Across diverse clinical tasks, our approach improves predictive performance, robustness under missing data, and uncertainty calibration relative to strong baselines. These results suggest that modeling what is not observed rather than only what is constitutes a critical inductive bias for healthcare foundation models.