Graph neural network modelling as a potentially effective method for predicting and analyzing procedures based on patients' diagnoses
作者：
Highlights：
• Graph Neural Networks applied to the analysis of ICD ontologies in patient populations
• Representation of Patient Populations in Graph structures
• Classification of medical procedures and patient clustering based on similar procedures using graph methods
• High accuracy of GNN methods compared to baseline models
• Applicability of method for recommender systems as well as empirical analysis of use of procedures and related ICDs within a patient population
摘要
•Graph Neural Networks applied to the analysis of ICD ontologies in patient populations•Representation of Patient Populations in Graph structures•Classification of medical procedures and patient clustering based on similar procedures using graph methods•High accuracy of GNN methods compared to baseline models•Applicability of method for recommender systems as well as empirical analysis of use of procedures and related ICDs within a patient population
论文关键词：Graph neural networks,Recommender systems,Diagnoses,Medical procedures
论文评审过程：Received 14 December 2021, Revised 29 June 2022, Accepted 9 July 2022, Available online 19 July 2022, Version of Record 5 August 2022.
论文官网地址：https://doi.org/10.1016/j.artmed.2022.102359
