Natural Language Processing (NLP) Clinical Extraction Method

Gawron A, Gundlapalli AV, Gawron LM, Samore MH, Rubin MA, Jones MM, Weir CR. Natural Language Processing (NLP) Clinical Extraction Method. 2017 Jul 1.

Related HSR&D Project(s)

• CDA 10-030 – Actionable Knowledge to Guide Antibiotic Stewardship
• CRE 12-320 – Veterans Like Mine Cognitive Support for Therapeutic Decision Making
• CRE 12-321 – Cognitive Support Informatics for Nurse Medication Stewardship
• IIR 12-084 – Current Evidence and Early Warning Indicators of Homelessness Risk Among Veterans

This method was developed by IDEAS researchers, validated using VA records, and tested using a number of Use Case scenarios. VA examples include: detecting suspected pneumonia cases, extracting social determinants of health in cardiac cases (PI: Salomeh Keyhani MD, San Francisco VAMC); detecting patients exhibiting significant carotid stenosis (PI: Salomeh Keyhani MD, San Francisco VAMC); encoding the exam qualities and findings from colonoscopy patients (PI: Andrew Gawron MD PhD, Salt Lake City VAMC, gastroenterology); duration of pain in surgical patients (PI: Luke Martin MD, Salt Lake City VAMC, surgery); identifying contraceptive usage in high risk pregnancy populations (PI: Lori Gawron MD PhD, Salt Lake City VAMC, Obstetrics and Gynecology); and identifying reasons for exam to reduce low-value exam ordering practices within the VA (PI: Salomeh Keyhani MD, San Francisco VAMC; VA pilot-funded project)