aiqbench analysis
SynEHR introduces a lightweight, LoRA-adapted framework that effectively synthesizes longitudinal electronic health records by jointly modeling irregular temporal evolution and intra-visit clinical structures. This approach addresses critical gaps in current generative models, producing sequences that are both temporally faithful and clinically coherent for downstream analysis. By leveraging large language models with parameter-efficient fine-tuning, the system offers a scalable solution for generating synthetic patient data where real records are scarce or privacy-restricted.
SynEHR: Joint Modeling Inter-visit Temporal Evolution and Intra-visit Clinical Structure for Longitudinal EHR Synthesis
arXiv:2608.21673v1 Announce Type: cross Abstract: Longitudinal electronic health records (EHRs) document patients' sequences of clinical visits over time, preserving the temporal evolution of disease progression and care delivery. However, real longitudinal EHRs are difficult to access because they contain large amounts of fine-grained, patient-specific information. Synthetic EHR generation therefore provides a valuable approach for preserving the statistical patterns and clinical structure of patient visit trajectories, enabling broader modeling and analysis when real records are limited. Although recent generative models have made progress in producing future visit sequences, they remain limited in explicitly integrating inter-visit irregular temporal evolution and intra-visit clinical event structures in EHRs, leading to clinically inconsistent and temporally unrealistic visit sequences. In this work, we propose SynEHR, a lightweight adaptive LLM-based framework for longitudinal EHR synthesis. There are two novel designs in SynEHR, i.e., a Temporal State Conditioning Module captures irregular temporal states across visits and a Temporal-Relational Adaptation Module combines these states with patient history to dynamically construct patient-specific relational representations. SynEHR then builds on a parameter-efficient LoRA-adapted language-model generator with next-visit generation capability to train the two modules for temporally and clinically informed generation. Extensive experiments on real-world EHR datasets across fidelity, privacy, and downstream utility evaluations demonstrate that SynEHR outperforms state-of-the-art models by generating more clinically coherent and temporally faithful longitudinal EHR data.
Abstract:Longitudinal electronic health records (EHRs) document patients' sequences of clinical visits over time, preserving the temporal evolution of disease progression and care delivery. However, real longitudinal EHRs are difficult to access because they contain large amounts of fine-grained, patient-specific information. Synthetic EHR generation therefore provides a valuable approach for preserving the statistical patterns and clinical structure of patient visit trajectories, enabling broader modeling and analysis when real records are limited. Although recent generative models have made progress in producing future visit sequences, they remain limited in explicitly integrating inter-visit irregular temporal evolution and intra-visit clinical event structures in EHRs, leading to clinically inconsistent and temporally unrealistic visit sequences. In this work, we propose SynEHR, a lightweight adaptive LLM-based framework for longitudinal EHR synthesis. There are two novel designs in SynEHR, i.e., a Temporal State Conditioning Module captures irregular temporal states across visits and a Temporal-Relational Adaptation Module combines these states with patient history to dynamically construct patient-specific relational representations. SynEHR then builds on a parameter-efficient LoRA-adapted language-model generator with next-visit generation capability to train the two modules for temporally and clinically informed generation. Extensive experiments on real-world EHR datasets across fidelity, privacy, and downstream utility evaluations demonstrate that SynEHR outperforms state-of-the-art models by generating more clinically coherent and temporally faithful longitudinal EHR data.
12 pages. CIKM '26
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2608.21673 [cs.LG]
(or arXiv:2608.21673v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2608.21673 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
https://doi.org/10.1145/3799682.3840734 Focus to learn more DOI(s) linking to related resources
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 324 words
Related coverage
TrustFormer: Cross-Temporal and Cross- Dimensional Transformer for Task-Specific Multi-Dimensional Trust Evaluation
arXiv · Aug 27, 2026
STRIDE: Spatial-Temporal Representation for Interval-conditioned Disease Evolution in Longitudinal Glioblastoma MRI
arXiv · Oct 8, 2026
CipherSight: Robust Website Fingerprinting via Record-Resource Semantic Supervision under Distribution Shifts
arXiv · Aug 17, 2026
Securing Time Integrity in Energy IoT Against Clock Drift and Y2K38 Failures
arXiv · Sep 1, 2026