"Towards An AI-Powered Application for Laboratory Test Results Comprehension "
Wednesday, Oct 02, 2024
Abstract:
Viewing laboratory test results is the most frequent activity for patients accessing patient portals, yet the interpretation of these results can be challenging and confusing. While previous research has explored various methods of presenting lab results, there has been limited focus on providing tailored information support based on an individual's medical context. Our AHRQ-funded LabGenie project addresses this gap by developing a user-centered, web-based tool designed to enhance patient engagement and understanding of lab results, particularly for older adults with multiple chronic conditions who often face difficulties due to limited health literacy and technology skills. In this talk, I will present a series of studies that investigate the challenges older adults face in understanding lab test results and our efforts to leverage generative AI and advanced biostatistical methods to provide better support. Our formative research included a survey of 270 patients to identify factors affecting their comprehension of lab results. Additionally, we evaluated the effectiveness of four large language models (LLMs) for answering lab-related questions sourced from a community Q&A website. To further enhance the accuracy of lab test interpretation, we analyzed the seasonality of lab results in a cohort of older adults with Alzheimer’s disease. We also used knowledge-augmented LLMs to determine normal ranges of various condition, predict differential diagnoses based on lab results and other medical information, and generate tailored questions for patients to follow with their doctors.
Through these studies, we aim to develop innovative, AI-driven approaches that improve the accessibility and comprehension of lab results for older adults, ultimately promoting better patient engagement and shared decision-making in healthcare.
Speaker's Bio:
Dr. Zhe He is Interim Director for the Institute for Successful Longevity and an Associate Professor in the School of Information (iSchool) at Florida State University. He is also Director of Biostatistics, Informatics, and Research Design (BIRD) Program in the UF-FSU Clinical and Translational Science Award Hub. His research expertise includes machine learning, natural language processing, and knowledge representation. The overarching goal of his research is to improve population health and advance biomedical research through the application of informatics and data science. His research has been funded by National Institute on Aging, National Library of Medicine, National Institute of Mental Health, and Agency for Healthcare Research and Quality.
