SynEHR: Joint Modeling Inter-visit Temporal Evolution and Intra-visit Clinical Structure for Longitudinal EHR Synthesis
Researchers have developed a new framework called SynEHR for synthesizing electronic health records (EHRs). This framework is designed to capture the temporal evolution of disease progression and care delivery in EHRs. Unlike previous models, SynEHR explicitly integrates inter-visit irregular temporal evolution and intra-visit clinical event structures. The authors propose two novel designs: a Temporal State Conditioning Module and a Temporal-Relational Adaptation Module. The
Researchers have developed a new framework called SynEHR for synthesizing electronic health records (EHRs). This framework is designed to capture the temporal evolution of disease progression and care delivery in EHRs. Unlike previous models, SynEHR explicitly integrates inter-visit irregular temporal evolution and intra-visit clinical event structures. The authors propose two novel designs: a Temporal State Conditioning Module and a Temporal-Relational Adaptation Module. These modules are used to generate more clinically coherent and temporally faithful longitudinal EHR data. Experiments on real-world datasets demonstrate that SynEHR outperforms state-of-the-art models in terms of fidelity, privacy, and downstream utility.
---
Why it matters: This matters because it can help improve the accuracy and reliability of AI-powered healthcare analytics by generating more realistic and clinically relevant EHRs.
Source: https://arxiv.org/abs/2608.21673
This article was originally published at: https://arxiv.org/abs/2608.21673