Reproduce “PatientEG Dataset: Bringing Event Graph Model with Temporal Relations to Electronic Medical Records” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Xuli Liu, Jihao Jin, Qi Wang, Tong Ruan, Yangming Zhou, Daqi Gao, Yichao Yin
The PatientEG dataset introduces an event graph model for Electronic Medical Records (EMRs) that explicitly represents medical activities and their temporal relations, enhancing data semantics and query efficiency. It transforms EMRs from Shanghai Shuguang Hospital into over 1.1 million RDF triples and links a substantial portion of entities to a biomedical knowledge graph for standardization.
Key insights
- Explicitly modeling temporal relationships between medical events within an EMR graph structure significantly improves the data's utility for complex clinical research queries.
- Linking non-standardized EMR entity values to an external biomedical knowledge graph through instance matching effectively normalizes data and enriches its semantic content.
- An event-centric graph model provides a semantically rich and query-efficient alternative to traditional relational databases, facilitating advanced EMR analysis and knowledge discovery.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.