# mathttMedGraph: Structural and Temporal Representation Learning of Electronic Medical Records

arXiv: https://arxiv.org/abs/1912.03703 · HTML: https://mlsanity.com/paper/1912.03703/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Dec 2019.

## Built on by (most cited first, 10 of 10)

- [Graph Neural Networks for Tabular Data Learning: A Survey with Taxonomy and Directions](https://mlsanity.com/paper/2401.02143/) (2024)
- [Graph-Text Multi-Modal Pre-training for Medical Representation Learning](https://mlsanity.com/paper/2203.09994/) (2022)
- [TabGSL: Graph Structure Learning for Tabular Data Prediction](https://mlsanity.com/paper/2305.15843/) (2023)
- [Graph AI in Medicine](https://mlsanity.com/paper/2310.13767/) (2023)
- [The Taxonomies, Training, and Applications of Event Stream Modelling for Electronic Health Records](https://mlsanity.com/paper/2603.14003/) (2026)
- [Demographic Aware Probabilistic Medical Knowledge Graph Embeddings of Electronic Medical Records](https://mlsanity.com/paper/2103.11951/) (2021)
- [SWoTTeD: An Extension of Tensor Decomposition to Temporal Phenotyping](https://mlsanity.com/paper/2310.01201/) (2023)
- [AdaMedGraph: Adaboosting Graph Neural Networks for Personalized Medicine](https://mlsanity.com/paper/2311.14304/) (2023)
- [Temporal Cascade and Structural Modelling of EHRs for Granular Readmission Prediction](https://mlsanity.com/paper/2102.02586/) (2021)
- [Towards modeling evolving longitudinal health trajectories with a transformer-based deep learning model](https://mlsanity.com/paper/2412.08873/) (2024)

## Abstract (arXiv)

Electronic medical record (EMR) data contains historical sequences of visits of patients, and each visit contains rich information, such as patient demographics, hospital utilisation and medical codes, including diagnosis, procedure and medication codes. Most existing EMR embedding methods capture visit-code associations by constructing input visit representations as binary vectors with a static vocabulary of medical codes. With this limited representation, they fail in encapsulating rich attribute information of visits (demographics and utilisation information) and/or codes (e.g., medical code descriptions). Furthermore, current work considers visits of the same patient as discrete-time events and ignores time gaps between them. However, the time gaps between visits depict dynamics of the patient's medical history inducing varying influences on future visits. To address these limitations, we present mathttMedGraph, a supervised EMR embedding method that captures two types of information: (1) the visit-code associations in an attributed bipartite graph, and (2) the temporal sequencing of visits through a point process. mathttMedGraph produces Gaussian embeddings for visits and codes to model the uncertainty. We evaluate the performance of mathttMedGraph through an extensive experimental study and show that mathttMedGraph outperforms state-of-the-art EMR embedding methods in several medical risk prediction tasks.