Clinical phenotyping in selected national networks: demonstrating the need for high-throughput, portable, and computational methods
作者：
Highlights：
• Computational methods for the identification of clinical phenotypes from EHR data are needed to advance our understanding of disease, treatments, and drug response.
• Machine learning approaches that rely on data patterns require fewer clinical domain experts and resources, and meet the growing demand for scalable and portable phenotyping tools.
• Research networks and phenotype developers should support collaboration and data standards that will enable computational phenotyping derived from data rather than experts.
摘要
•Computational methods for the identification of clinical phenotypes from EHR data are needed to advance our understanding of disease, treatments, and drug response.•Machine learning approaches that rely on data patterns require fewer clinical domain experts and resources, and meet the growing demand for scalable and portable phenotyping tools.•Research networks and phenotype developers should support collaboration and data standards that will enable computational phenotyping derived from data rather than experts.
论文关键词：Machine learning,Clinical phenotyping,Electronic health records,Networked research,Precision medicine
论文评审过程：Received 22 February 2016, Accepted 30 May 2016, Available online 25 June 2016, Version of Record 11 July 2016.
论文官网地址：https://doi.org/10.1016/j.artmed.2016.05.005
