Feature engineering and selection for rheumatoid arthritis disease activity classification using electronic medical records
Published in ICML Workshop on Machine Learning for Clinical Data Analysis, 2012
Recommended citation: Chen Lin, Helena Canhao, Timothy Miller, Dmitriy Dligach, Robert M Plenge, Elizabeth W Karlson, and Guergana K Savova. 2012. Feature engineering and selection for rheumatoid arthritis disease activity classification using electronic medical records. In ICML Workshop on Machine Learning for Clinical Data Analysis. https://people.cs.pitt.edu/~milos/icml_clinicaldata_2012/Papers/Oral_Chen_Guergana_ICML_Clinical_2012.pdf
Abstract:
Abstract We study feature engineering and feature selection related to a clinical research application–automatically discovering the patient’s disease activity from the electronic medical records. Different feature representations of clinical documents such as user specified terms, Unified Medical Language System Concept Unique Identifiers, bag of words, and bigram features are compared with filter-based feature selection methods. Performance evaluations are conducted given all feature sets and under varied feature selection conditions on a gold standard set.