Short bio: Ping Wang is a Ph.D. candidate in the Department of Computer Science at Virginia Tech, supervised by Dr. Chandan K. Reddy. She received her M.S. degree in Computer Science from Wayne State University in 2016. Her research focuses on question answering, graph mining, information extraction and survival analysis with their applications in the healthcare domain. She has published papers in leading conferences and high-impact journals, including WWW, CIKM, AAAI, NAACL, TKDE and ACM Computing Surveys (CSUR). She has served as the reviewer for conferences and journals in machine learning, NLP, and healthcare communities, including ACL, EMNLP, COLING, AMIA, ACM HEALTH, CSUR, TKDD and TKDE.
Title: Automatic Question Answering and Knowledge Discovery on Electronic Health Records
Abstract: Healthcare systems are changing in the era of big data. Advances in artificial intelligence and digitization in healthcare have allowed healthcare providers to effectively sift through tremendous amounts of information, which eventually guide them to better take care of their patients. There are various types of health information ranging from medical literature to pathology reports. My research goal is to develop machine learning methods that can efficiently utilize Electronic Health Records (EHRs), which contain medical and treatment history of patients, to facilitate decision making of physicians in their clinical practice.
In this talk, I will first introduce my recent work on question answering for medical records including text-to-SQL query generation and clinical knowledge base question answering, which aims to seek answers from structured EHRs and unstructured clinical notes for comprehending critical information from massive medical repositories. Then, I will present a self-supervised learning approach of contextual embeddings for link prediction in heterogeneous networks, which can be used to discover correlations of different entities (e.g., diseases, medications, and treatments) in tabular EHRs. Finally, I will share my vision on future research topics that I plan to pursue.
Monday, March 8, 2021 at 11:15am to 12:15pmVirtual Event