Suicide presents a major public health challenge worldwide and affects people throughout their life. It demands immediate attention and a comprehensive understanding of the underlying suicide causes. The National Violent Death Reporting System (NVDRS) is a population-based active surveillance system that collects information on violent deaths that occurred among both residents and non-residents in the United States. It serves as a valuable repository of death investigation notes, offering crucial information and contexts surrounding suicide deaths, which are essential for developing NLP systems that can enhance our understanding of suicide causes.
The presenters discussed their work on effectively extracting suicide causes from death investigation notes using NLP approaches. Through analysis, they noticed a significant performance mismatch between different states. They will argue that the inherent data annotation inconsistencies exist in NVDRS between different states and even within a single state, which can be one of the main causes of the observed performance gap. This reveals an unmet need for approaches to identify and address the annotation inconsistencies in death investigation notes.
The presenters also described their NLP approach designed to explore the data annotation inconsistencies in NVDRS death investigation notes, how they identified problematic data instances that may contribute to these in consistencies, and further verify the effectiveness of label correction.
Watch the Recording
Presenters
Song Wang is a 2nd-year PhD student of Electrical and Computer Engineering at UT Austin, supervised by Dr. Joydeep Ghosh, Dr. Ying Ding, and Dr. Yifan Peng from Weill Cornell Medicine. His research focuses on Natural Language Processing, Multi-Modal Representation Learning. He completed his M.S. in Electrical and Computer Engineering at UT Austin in 2021.
Yifan Peng, PhD, is an Assistant Professor in the Division of Health Sciences Department of Population Health Sciences at Weill Cornell Medicine. His main research interests include BioNLP and medical image analysis. He has published in major AI and healthcare informatics venues, including ACL, CVPR, MICCAI, and ICHI, as well as medical venues including Nucleic Acids Research, npj Digital Medicine, and Ophthalmology Science. His research has been funded by federal agencies, including NIH and NSF and industries such as Amazon and Google. He is an Editorial Board Member for the Journal of Biomedical Informatics.