mlsanityData to 14 Oct 2026

Latent Factor Point Processes for Patient Representation in Electronic Health Records

arXiv 2508.20327 ↗First posted 28 Aug 2025stat.ME stat.MLcited by 0

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 39 of its 100 references that are in the map. First posted on arXiv 28 Aug 2025.

Builds on

39 of 100 references.

References · 39

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2020Med-BERT: pre-trained contextualized embeddings on large-scale structured electronic health records for disease predictioncited by 373
  3. 2019BEHRT: Transformer for Electronic Health Recordscited by 263
  4. 2022GatorTron: A Large Clinical Language Model to Unlock Patient Information from Unstructured Electronic Health Recordscited by 238
  5. 2022Large Language Models are Few-Shot Clinical Information Extractorscited by 207
  6. 2020Language Models Are An Effective Patient Representation Learning Technique For Electronic Health Record Datacited by 87
  7. 2020Deep Representation Learning of Patient Data from Electronic Health Records (EHR): A Systematic Reviewcited by 47
  8. 2018Clinical Concept Embeddings Learned from Massive Sources of Multimodal Medical Datacited by 35
  9. 2019MetaPred: Meta-Learning for Clinical Risk Prediction with Limited Patient Electronic Health Recordscited by 33
  10. 2021Federated Causal Inference in Heterogeneous Observational Datacited by 24
  11. 2015Sparse and low-rank multivariate Hawkes processescited by 19
  12. 2017Learning from Clinical Judgments: Semi-Markov-Modulated Marked Hawkes Processes for Risk Prognosiscited by 18
27 more
  1. 2019Estimating Number of Factors by Adjusted Eigenvalues Thresholdingcited by 14
  2. 2021Machine learning for modeling the progression of Alzheimer disease dementia using clinical data: a systematic literature reviewcited by 13
  3. 2022Federated Offline Reinforcement Learningcited by 13
  4. 2020Latent Network Structure Learning from High Dimensional Multivariate Point Processescited by 12
  5. 2022Semi-supervised Triply Robust Inductive Transfer Learningcited by 12
  6. 2019Identifying Sub-Phenotypes of Acute Kidney Injury using Structured and Unstructured Electronic Health Record Data with Memory Networkscited by 10
  7. 2021Multi-source Learning via Completion of Block-wise Overlapping Noisy Matricescited by 10
  8. 2017The Multivariate Hawkes Process in High Dimensions: Beyond Mutual Excitationcited by 9
  9. 2017Semi-Supervised Approaches to Efficient Evaluation of Model Prediction Performancecited by 9
  10. 2021Surrogate Assisted Semi-supervised Inference for High Dimensional Risk Predictioncited by 6
  11. 2021Maximum Likelihood Estimation for Hawkes Processes with self-excitation or inhibitioncited by 5
  12. 2023Inference of Dependency Knowledge Graph for Electronic Health Recordscited by 5
  13. 2021Double Robust Semi-Supervised Inference for the Mean: Selection Bias under MAR Labeling with Decaying Overlapcited by 4
  14. 2023Guided Adversarial Robust Transfer Learning with Source Mixingcited by 4
  15. 2020Multivariate Temporal Point Process Regressioncited by 3
  16. 2022Deep Stable Representation Learning on Electronic Health Recordscited by 3
  17. 2023Knowledge Graph Embedding with Electronic Health Records Data via Latent Graphical Block Modelcited by 3
  18. 2025Time-Aware Attention for Enhanced Electronic Health Records Modelingcited by 3
  19. 2020Prior Adaptive Semi-supervised Learning with Application to EHR Phenotypingcited by 2
  20. 2024Efficient collaborative learning of the average treatment effectcited by 2
  21. 2025Balancing Interpretability and Flexibility in Modeling Diagnostic Trajectories with an Embedded Neural Hawkes Process Modelcited by 2
  22. 2021Causal Discovery in High-Dimensional Point Process Networks with Hidden Nodescited by 1
  23. 2022Assessing the Most Vulnerable Subgroup to Type II Diabetes Associated with Statin Usage: Evidence from Electronic Health Record Datacited by 1
  24. 2024Regularised Spectral Estimation for High-Dimensional Point Processescited by 1
  25. 2024Spectral Differential Network Analysis for High-Dimensional Time Seriescited by 1
  26. 2025Fast and robust invariant generalized linear modelscited by 1
  27. 2025Semi-supervised Clustering Through Representation Learning of Large-scale EHR Datacited by 1

Built on by

Not cited yet.