ABOUT THE LECTURE:
Clinical AI is only as reliable as the data it is built on, yet the validity of these inputs are often overlooked in real-world practice. Validated input data and deliberate deployment infrastructure must be preconditions for trustworthy clinical AI rather than afterthoughts. Join Dr Ian Wong as he shares two complementary lines of work that address this gap. The OPTIC study, a large prospective collaboration across Duke and Emory, investigates pulse oximetry bias and the risk of hidden hypoxemia linked to differences in skin tone, with clear implications for patient safety and health equity that are highly relevant to our region. Alongside this, an LLM based framework demonstrates how clinically important signals buried within unstructured notes can be extracted and translated into actionable insights, with a focus on enhancing risk stratification and care pathways in pulmonary embolism.
HOST:
Dr Amanda Lam
Department of Data Science
Singapore General Hospital
VENUE:
Duke-NUS Medical School
Amphitheatre, Level 2
CONTACT PERSON:
Ms Kathleen Chan (kathleen.chan@duke-nus.edu.sg)
Duke-NUS Research Affairs Department
Date and Time
17 Apr 2026 @ 12:00 - 17 Apr 2026 @ 13:00
Speaker
Dr An-Kwok Ian Wong
Assistant Professor
Departments of Medicine and Biostatistics & Bioinformatics
Duke University School of Medicine
A. Ian Wong, MD, PhD is an Assistant Professor of Medicine at Duke University, board-certified in Pulmonary, Critical Care, and Clinical Informatics. His research focuses on trustworthy AI for critical illness and pulmonary disease, with NIH-funded work on pulse oximetry bias (OPTIC) and LLM-based clinical data extraction. He practices at the Durham VA Medical Center, Duke University Hospital, and co-chairs SCCM's Critical Care Innovation Incubator.