Latent Factor Point Processes for Patient Representation in Electronic Health Records

2025-08-27

AI Summary

Clinical codes in electronic health records (EHRs) exhibit irregular temporal structures, yet existing methods often oversimplify them as static counts or rely on strong parametric assumptions, limiting their ability to capture patient-level dynamic heterogeneity. To address this, we propose a latent factor point process model that employs low-dimensional latent Poisson processes to govern the intensity of high-dimensional clinical code occurrences, explicitly modeling underlying disease progression. We innovatively incorporate Fourier feature embeddings to capture inter-subgroup temporal pattern variations and construct frequency-domain patient representations via spectral density matrix analysis, enabling efficient high-dimensional statistical inference. Evaluated on an Alzheimer’s disease cohort, our method significantly improves both supervised disease classification and unsupervised patient clustering performance, while uncovering clinically meaningful patient subgroups reflecting heterogeneous disease trajectories.

Technology Category

• Machine Learning: Probabilistic Circuits and Graphical Models
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

Electronic health records (EHR) contain valuable longitudinal patient-level information, yet most statistical methods reduce the irregular timing of EHR codes into simple counts, thereby discarding rich temporal structure. Existing temporal models often impose restrictive parametric assumptions or are tailored to code level rather than patient-level tasks. We propose the latent factor point process model, which represents code occurrences as a high-dimensional point process whose conditional intensity is driven by a low dimensional latent Poisson process. This low-rank structure reflects the clinical reality that thousands of codes are governed by a small number of underlying disease processes, while enabling statistically efficient estimation in high dimensions. Building on this model, we introduce the Fourier-Eigen embedding, a patient representation constructed from the spectral density matrix of the observed process. We establish theoretical guarantees showing that these embeddings efficiently capture subgroup-specific temporal patterns for downstream classification and clustering. Simulations and an application to an Alzheimer's disease EHR cohort demonstrate the practical advantages of our approach in uncovering clinically meaningful heterogeneity.

Problem

• Modeling irregular EHR timing beyond simple counts
• Capturing patient-level temporal patterns for clinical tasks
• Uncovering disease heterogeneity through latent process representation

Innovation

• Latent factor point process model
• Low-rank latent Poisson process structure
• Fourier-Eigen patient spectral embeddings