Utilizing large Language Models (LLM) to Optimize Domain-Specific Natural Language Processing (NLP) for Identifying Patients with No Reason for Not Prescribing ACEI/ARB in Chronic Kidney Disease (CKD) Management
Poster Number: P05
Presentation Time: 05:00 PM - 06:30 PM
Abstract Keywords: Natural Language Processing, Large Language Models (LLMs), Information Extraction
Primary Track: Applications
Programmatic Theme: Academic Informatics / LIEAF
This study developed a framework utilizing nurse-annotated examples and LLM-generated training data to improve NLP models for identifying reasons for not prescribing ACEI/ARB in CKD patients. Combining nurse-annotated and LLM-generated data resulted in the best model performance, addressing the challenge of data annotation in healthcare NLP applications. This approach is important for developing accurate, efficient, and small NLP models tailored to specific healthcare domains while reducing the burden of manual data annotation.
Speaker(s):
Mohammed Al-Garadi, PhD
VUMC
Author(s):
Mohammed Al-Garadi, PhD - VUMC; Michael Matheny, MD, MS, MPH, FACMI, FAMIA - Vanderbilt University Medical Center; Sankar Navaneethan, M.D - Section of Nephrology, Michael E. DeBakey Veterans Affairs Medical Center, Houston, TX; Richard Noriega, BSC - VUMC; Dax Westerman, MS - Vanderbilt University Medical Center Department of Biomedical Informatics; Parker Gregg, M.D - Section of Nephrology, Michael E. DeBakey Veterans Affairs Medical Center, Houston, TX; Sheena Wydermyer, M.D - Section of Nephrology, Michael E. DeBakey Veterans Affairs Medical Center, Houston, TX; Jennifer Arney, PhD - Michael E. DeBakey Veterans Affairs Medical Center, Houston; Salim Virani, M.D - Michael E. DeBakey Veterans Affairs Medical Center, Houston, TX; Michael Herrera, BSC - Michael E. DeBakey Veterans Affairs Medical Center, Houston, TX; Peter Richardson, BCS - Michael E. DeBakey Veterans Affairs Medical Center, Houston, TX; Jill Whitaker, Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN - Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN; Tina French, RN, CPHQ - Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN,; Lisa Roddy, RN - VUMC; Glenn Gobbel - VA Tennessee Valley Healthcare System;
Poster Number: P05
Presentation Time: 05:00 PM - 06:30 PM
Abstract Keywords: Natural Language Processing, Large Language Models (LLMs), Information Extraction
Primary Track: Applications
Programmatic Theme: Academic Informatics / LIEAF
This study developed a framework utilizing nurse-annotated examples and LLM-generated training data to improve NLP models for identifying reasons for not prescribing ACEI/ARB in CKD patients. Combining nurse-annotated and LLM-generated data resulted in the best model performance, addressing the challenge of data annotation in healthcare NLP applications. This approach is important for developing accurate, efficient, and small NLP models tailored to specific healthcare domains while reducing the burden of manual data annotation.
Speaker(s):
Mohammed Al-Garadi, PhD
VUMC
Author(s):
Mohammed Al-Garadi, PhD - VUMC; Michael Matheny, MD, MS, MPH, FACMI, FAMIA - Vanderbilt University Medical Center; Sankar Navaneethan, M.D - Section of Nephrology, Michael E. DeBakey Veterans Affairs Medical Center, Houston, TX; Richard Noriega, BSC - VUMC; Dax Westerman, MS - Vanderbilt University Medical Center Department of Biomedical Informatics; Parker Gregg, M.D - Section of Nephrology, Michael E. DeBakey Veterans Affairs Medical Center, Houston, TX; Sheena Wydermyer, M.D - Section of Nephrology, Michael E. DeBakey Veterans Affairs Medical Center, Houston, TX; Jennifer Arney, PhD - Michael E. DeBakey Veterans Affairs Medical Center, Houston; Salim Virani, M.D - Michael E. DeBakey Veterans Affairs Medical Center, Houston, TX; Michael Herrera, BSC - Michael E. DeBakey Veterans Affairs Medical Center, Houston, TX; Peter Richardson, BCS - Michael E. DeBakey Veterans Affairs Medical Center, Houston, TX; Jill Whitaker, Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN - Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN; Tina French, RN, CPHQ - Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN,; Lisa Roddy, RN - VUMC; Glenn Gobbel - VA Tennessee Valley Healthcare System;
Utilizing large Language Models (LLM) to Optimize Domain-Specific Natural Language Processing (NLP) for Identifying Patients with No Reason for Not Prescribing ACEI/ARB in Chronic Kidney Disease (CKD) Management
Category
Poster - Regular
Description
Date: Tuesday (11/12)
Time: 05:00 PM to 06:30 PM
Room: Grand Ballroom (Posters)
Time: 05:00 PM to 06:30 PM
Room: Grand Ballroom (Posters)