Prediction of Survival Outcomes under Clinical Presence Shift: A Joint Neural Network Architecture

2025-08-07

AI Summary

Clinical prediction models often suffer significant performance degradation when deployed across institutions due to neglect of “clinical presence”—the spatiotemporal distribution characteristics of clinical observations—particularly under clinical presence shift: systematic inter-hospital or regional differences in observation frequency, patterns, and missingness mechanisms. This work formally defines clinical presence shift for the first time and proposes a novel multi-task recurrent neural network that jointly models observation inter-arrival times, missingness mechanisms, and survival outcomes. By end-to-end learning latent regularities in clinical observation behavior, the model explicitly mitigates generalization bottlenecks arising from heterogeneous data collection practices. Experiments on MIMIC-III demonstrate statistically significant improvements over state-of-the-art survival prediction models in both C-index and Brier score, alongside substantially enhanced cross-center transferability. This approach establishes a new paradigm for developing robust, deployable clinical AI systems.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior
• Multiagent Systems: Modeling other Agents

Application Category

• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Electronic health records arise from the complex interaction between patients and the healthcare system. This observation process of interactions, referred to as clinical presence, often impacts observed outcomes. When using electronic health records to develop clinical prediction models, it is standard practice to overlook clinical presence, impacting performance and limiting the transportability of models when this interaction evolves. We propose a multi-task recurrent neural network that jointly models the inter-observation time and the missingness processes characterising this interaction in parallel to the survival outcome of interest. Our work formalises the concept of clinical presence shift when the prediction model is deployed in new settings (e.g. different hospitals, regions or countries), and we theoretically justify why the proposed joint modelling can improve transportability under changes in clinical presence. We demonstrate, in a real-world mortality prediction task in the MIMIC-III dataset, how the proposed strategy improves performance and transportability compared to state-of-the-art prediction models that do not incorporate the observation process. These results emphasise the importance of leveraging clinical presence to improve performance and create more transportable clinical prediction models.

Problem

• Model survival outcomes considering clinical presence shift
• Improve transportability of prediction models across settings
• Address missing data and observation time in EHRs

Innovation

• Joint neural network models clinical presence
• Multi-task RNN handles missingness and survival
• Improves transportability across clinical settings