Illinois Medicine

Illinois Medicine | 23 prevent children at risk of ROP from going blind," Chan says. Through a partnership with the Department of Community and Family Medicine, as well as the Institute for Healthcare Delivery Design at UI Health, the Department of Ophthalmology and Visual Sciences is now deploying the Eyenuk program for diabetic patients. This program, which harnesses the power of AI to screen for diabetic retinopathy, will improve access to care for diabetic patients in the communities served by UI Health and has the potential to improve visual outcomes in this patient population. "Developing and implementing AI in clinical care settings is complex, and it is especially important that we make sure that the data initially collected is good and relevant to the conditions and questions that we are trying to address," Chan notes. "The output is only as good as the input, so it's imperative that we have the correct metrics in order to collect accurate data that will yield appropriate and meaningful results." Investigators in the Department of Ophthalmology and Visual Sciences are collaborating closely with faculty from the Department of Biomedical Engineering to develop AI applications that can be used with various imaging systems to detect the presence and progression of eye diseases. For instance, Jennifer I. Lim, MD, FARVO, is doing preliminary work using AI to look at biomarkers and imaging systems such as ocular optical coherence tomography (OCT) and OCT with angiography to determine whether biomarkers in the eye can help differentiate stages of sickle cell eye disease. Lim works closely with Xincheng Yao, PhD, the Richard and Loan Hill Professor in the Department T Spearheaded by department of ophthalmology head R.V. Paul Chan, MD, MSc, MBA; adjunct assistant professor of ophthalmology Joelle A. Hallak, MS '08, PhD '15; and assistant professor of ophthalmology Darvin Yi, PhD, the Ai-O Program is dedicated to building a methodology that analyzes data to address bias and create robust automated systems that represent diverse populations. It aims to develop human-centered AI for patient care in an interpretable, ethical and reproducible manner. When it came time to launch CHEMA, the experienced faculty in the Department of Ophthalmology and Visual Sciences at UIC proved invaluable. "There's much potential and still much to learn with AI," says Chan, who co-directs the Ai-O Program with Yi. "We're already helping to level the disparities in healthcare. We want treatments that are applicable to all people, and we are working to validate technology in a way that is representative of the total population." As diagnostic tools that have the potential to predict outcomes, AI methods are being applied to different ophthalmic imaging techniques such as optical coherence tomography or color images of the retina or other parts of the eye. "AI for ophthalmology is here, and faculty at UIC are leading efforts in this field on a national and international level," says Chan. Chan has worked closely with collaborators from the Casey Eye Institute at Oregon Health & Sciences University, Massachusetts General Hospital at Harvard Medical School, and the Department of Ophthalmology at the University of Colorado to develop an AI algorithm for the diagnosis of retinopathy of prematurity (ROP). "In 2019, this AI algorithm received breakthrough status by the FDA, and aims to THE UIC ARTIFICIAL INTELLIGENCE IN OPHTHALMOLOGY PROGRAM (Ai-O), launched in 2020, preceded the development of the Center for Health Equity using Machine Learning and Artificial Intelligence (CHEMA). The field of ophthalmology received one of the first FDA-approved algorithms in 2018 for clinical use to help detect diabetic retinopathy, a condition that can cause blindness. B Y C A R L A B E E C H E R P H O T O : D I A N E S M U T N Y

• Cover