Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

GraphCare: Enhancing Healthcare Predictions with Personalized Knowledge Graphs

About

Clinical predictive models often rely on patients' electronic health records (EHR), but integrating medical knowledge to enhance predictions and decision-making is challenging. This is because personalized predictions require personalized knowledge graphs (KGs), which are difficult to generate from patient EHR data. To address this, we propose \textsc{GraphCare}, an open-world framework that uses external KGs to improve EHR-based predictions. Our method extracts knowledge from large language models (LLMs) and external biomedical KGs to build patient-specific KGs, which are then used to train our proposed Bi-attention AugmenTed (BAT) graph neural network (GNN) for healthcare predictions. On two public datasets, MIMIC-III and MIMIC-IV, \textsc{GraphCare} surpasses baselines in four vital healthcare prediction tasks: mortality, readmission, length of stay (LOS), and drug recommendation. On MIMIC-III, it boosts AUROC by 17.6\% and 6.6\% for mortality and readmission, and F1-score by 7.9\% and 10.8\% for LOS and drug recommendation, respectively. Notably, \textsc{GraphCare} demonstrates a substantial edge in scenarios with limited data availability. Our findings highlight the potential of using external KGs in healthcare prediction tasks and demonstrate the promise of \textsc{GraphCare} in generating personalized KGs for promoting personalized medicine.

Pengcheng Jiang, Cao Xiao, Adam Cross, Jimeng Sun• 2023

Related benchmarks

TaskDatasetResultRank
Readmission predictionMIMIC IV
AUC-ROC0.685
90
Readmission predictionMIMIC-III (target)
AUPRC73.4
59
Mortality PredictionMIMIC-III
AUROC70.3
59
Mortality PredictionMIMIC IV
AUROC0.731
53
Medication RecommendationMIMIC-III
Jaccard Similarity49.8
42
Medication RecommendationMIMIC IV
Jaccard Similarity48.1
32
Length-of-Stay PredictionMIMIC-III--
28
Length-of-Stay PredictionMIMIC IV
Kappa29.8
9
Showing 8 of 8 rows

Other info

Follow for update