EMR-Based Phenotyping of Ischemic Stroke Using Supervised Machine Learning and Text Mining Techniques

• Published in: IEEE journal of biomedical and health informatics Vol. 24; no. 10; pp. 2922 - 2931
• Main Authors: Sung, Sheng-Feng, Lin, Chia-Yi, Hu, Ya-Han
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.10.2020; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Aged; Algorithms; Annotations; Classification; Classification algorithm; clinical diagnosis; Data mining; Data Mining - methods; Diagnosis, Computer-Assisted - methods; Diuretics; Electronic Health Records; electronic medical records; Female; Hospitals; Humans; Information management; Ischemia; Ischemic Stroke - classification; Ischemic Stroke - diagnosis; Learning algorithms; Machine learning; Machine learning algorithms; Male; Medical diagnosis; Medical records; Medical research; Natural Language Processing; Phenotypes; Phenotyping; Population studies; Stroke; Supervised Machine Learning; Task analysis; text mining; Unstructured data
• ISSN: 2168-2194; 2168-2208; 2168-2208
• DOI: 10.1109/JBHI.2020.2976931

Ischemic stroke is a major cause of death and disability in adulthood worldwide. Because it has highly heterogeneous phenotypes, phenotyping of ischemic stroke is an essential task for medical research and clinical prognostication. However, this task is not a trivial one when the study population is large. Phenotyping of ischemic stroke depends primarily on manual annotation of medical records in previous studies. This article evaluated various strategies for automated phenotyping of ischemic stroke into the four subtypes of the Oxfordshire Community Stroke Project classification based on structured and unstructured data from electronical medical records (EMRs). A total of 4640 adult patients who were hospitalized for acute ischemic stroke in a teaching hospital were included. In addition to the structured items in the National Institutes of Health Stroke Scale, unstructured clinical narratives were preprocessed using MetaMap to identify medical concepts, which were then encoded into feature vectors. Various supervised machine learning algorithms were used to build classifiers. The study results indicate that textual information from EMRs could facilitate phenotyping of ischemic stroke when this information was combined with structured information. Furthermore, decomposition of this multi-class problem into binary classification tasks followed by aggregation of classification results could improve the performance.