Latent Factor Point Processes for Patient Representation in Electronic Health Records

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Knight, Parker, Zhou, Doudou, Xia, Zongqi, Cai, Tianxi, Lu, Junwei
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908507603730432
author Knight, Parker
Zhou, Doudou
Xia, Zongqi
Cai, Tianxi
Lu, Junwei
author_facet Knight, Parker
Zhou, Doudou
Xia, Zongqi
Cai, Tianxi
Lu, Junwei
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2508_20327
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Latent Factor Point Processes for Patient Representation in Electronic Health Records
Knight, Parker
Zhou, Doudou
Xia, Zongqi
Cai, Tianxi
Lu, Junwei
Methodology
Machine Learning
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.
title Latent Factor Point Processes for Patient Representation in Electronic Health Records
topic Methodology
Machine Learning
url https://arxiv.org/abs/2508.20327