TrajSurv: Learning Continuous Latent Trajectories from Electronic Health Records for Trustworthy Survival Prediction

2025-08-01

AI Summary

This study addresses the dual challenges of modeling patient clinical progression and enabling trustworthy survival prediction from irregularly sampled longitudinal electronic health records (EHRs). We propose a continuous-time latent trajectory modeling framework based on neural controlled differential equations (NCDEs), integrated with time-aware contrastive learning to align latent representations with clinical states. A two-step interpretability mechanism dynamically links feature evolution to survival risk, while vector field learning combined with clustering enables clinical pathway pattern discovery and enhanced model transparency. Evaluated on MIMIC-III and eICU datasets, our method achieves competitive predictive accuracy and significantly outperforms existing deep survival models in model transparency and clinical interpretability—offering both principled temporal modeling and actionable, clinically grounded explanations.

Technology Category

• Machine Learning: Life-Long and Continual Learning
• Computer Vision: Interpretability, Explainability, and Transparency
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Trustworthy survival prediction is essential for clinical decision making. Longitudinal electronic health records (EHRs) provide a uniquely powerful opportunity for the prediction. However, it is challenging to accurately model the continuous clinical progression of patients underlying the irregularly sampled clinical features and to transparently link the progression to survival outcomes. To address these challenges, we develop TrajSurv, a model that learns continuous latent trajectories from longitudinal EHR data for trustworthy survival prediction. TrajSurv employs a neural controlled differential equation (NCDE) to extract continuous-time latent states from the irregularly sampled data, forming continuous latent trajectories. To ensure the latent trajectories reflect the clinical progression, TrajSurv aligns the latent state space with patient state space through a time-aware contrastive learning approach. To transparently link clinical progression to the survival outcome, TrajSurv uses latent trajectories in a two-step divide-and-conquer interpretation process. First, it explains how the changes in clinical features translate into the latent trajectory's evolution using a learned vector field. Second, it clusters these latent trajectories to identify key clinical progression patterns associated with different survival outcomes. Evaluations on two real-world medical datasets, MIMIC-III and eICU, show TrajSurv's competitive accuracy and superior transparency over existing deep learning methods.

Problem

• Model continuous clinical progression from irregular EHR data
• Align latent trajectories with actual patient states
• Link clinical progression patterns to survival outcomes

Innovation

• Uses NCDE for continuous latent trajectories
• Aligns latent and patient states contrastively
• Interprets trajectories via vector field and clustering