Natural language processing: applications in pediatric research

Published in Pediatric Biomedical Informatics: Computer Applications in Pediatric …, 2016, 2016

Recommended citation: Guergana Savova, John Pestian, Brian Connolly, Timothy Miller, Yizhao Ni, and Judith W Dexheimer. 2016. Natural language processing: applications in pediatric research. In Pediatric Biomedical Informatics: Computer Applications in Pediatric …, 2016.

Abstract:

We discuss specific biomedical Natural Language Processing-based applications that cover a wide spectrum of use cases within the field of translational and health services research. In our uses cases we focus on four categories of applications: (1) Information Extraction (IE), (2) Document Classification, (3) Patient Classification, and (4) Sentiment Analysis. We show how the extracted information could be used for (a) Phenotype identification, (b) Comparative effectiveness studies, (c) Cohort identification, (d) Meaningful Use, and (e) Linking patients’ phenotype and genotype. In addition, we discuss the use of Natural Language Processing components for de-identification of large collections of patient notes. We review the literature for examples of pediatric natural language processing applications and show the transferability of select adult clinical natural language processing applications to the pediatric population.