Chronic disease prediction using administrative data and graph theory: The case of type 2 diabetes
作者：
Highlights：
• Administrative healthcare data is used to identify high-risk chronic patients.
• Graph theory and social network analysis concepts were used to understand the disease progression.
• Prediction framework showed between 82% to 87% accuracy using different methods.
• Three predictive methods — regression, parameter optimization and tree classification were used.
• Binary tree classification showed higher performance compared to the other two.
摘要
•Administrative healthcare data is used to identify high-risk chronic patients.•Graph theory and social network analysis concepts were used to understand the disease progression.•Prediction framework showed between 82% to 87% accuracy using different methods.•Three predictive methods — regression, parameter optimization and tree classification were used.•Binary tree classification showed higher performance compared to the other two.
论文关键词：Disease prediction,Electronic medical records,Medical information systems,Network theory,Prediction theory,Type 2 diabetes
论文评审过程：Received 17 January 2019, Revised 18 April 2019, Accepted 28 May 2019, Available online 8 June 2019, Version of Record 28 June 2019.
论文官网地址：https://doi.org/10.1016/j.eswa.2019.05.048
