Research Assistant Professor, University of Texas, Health Science Center at Houston, School of Biomedical Informatics
“Unlocking Patient Information from Clinical Narratives.”
Thursday, March 2
Clinical and Translational Research Building
Room 2161, 1-2 p.m.
Read more on upcoming seminars from the Department of Health Outcomes & Policy here
As an important component of electronic health records (EHR) data, clinical narratives have been widely used for clinical, informatics and translational research. Accurate clinical studies often require detailed information associated with patients. However, much of the detailed patient information is locked in the narrative clinical text, which is not directly accessible for clinical applications that rely on structured data. Clinical Natural Language Processing (NLP) is the only way to unlock the rich patient information from clinical narratives to support more accurate clinical and translational studies. In this seminar, I will talk about unlocking patient information from clinical narratives using clinical NLP. More specifically, I will take the clinical abbreviations as an example to demonstrate how clinical NLP and machine learning methods could help handling clinical abbreviations to improve patient safety. Then, I will talk about several NLP topics such as Named Entity Recognition (NER), NLP applications on other medical text (e.g., clinical trials), and emerging clinical NLP technologies based on deep learning models.