Illinois Medicine

T Illinois Medicine | 25 B Y C A R L A B E E C H E R THE OPHTHALMOLOGY RESEARCH LABORATORY of Darvin Yi, PhD, assistant professor of ophthalmology, is dedicated to using AI technology to detect the tiniest indicators of eye abnormality and disease. After receiving a PhD in biomedical informatics from Stanford University, Yi joined the Department of Ophthalmology and Visual Sciences in July 2020 as the first AI scientist to work in the newly created Artificial Intelligence in Ophthalmology Program (Ai-O). His lab provides the data link between bench and bedside to help ophthalmologists prevent blindness and cure eye disease. A patient visits her primary care physician a year after giving birth. The doctor asks about her baby, and she shows him a photograph. Fortunately, the doctor is familiar with retinoblastoma, an eye cancer that begins in the retina and can spread to other parts of the body. When he sees its telltale unusual white reflection in the toddler's pupils, he refers her to an ophthalmologist. That intervention is lifesaving. Now imagine the ability to replicate that happenstance encounter ad infinitum. Such is the power of artificial intelligence. "The world is rapidly becoming more data-centric—especially in clinical medicine," he says. "Together with Dr. [R.V. Paul] Chan and all of the faculty of the Illinois Eye and Ear Infirmary, we hope to build an AI center that can take advantage of this trend." Yi's lab in the Lions of Illinois Eye Research Institute consists of three areas: clinical decision support, theoretical research and system innovation. He describes what he and his team of seven do as simply "making tools for doctors to use." "There's something physically useful about computers in that they are extremely scalable," says Yi. "You can buy 100 computers much faster than you can train 100 ophthalmologists, but there are still huge limits to what computers can do. They can't talk to patients. They can't perform surgery. They can't understand the nuances of care. Yet they do other things incredibly well." For instance, one of the projects under development in Yi's lab is the use of AI to predict ptosis, which causes eyelid drooping. This research project, called "AutoPtosis," has the potential to mass categorize human eyelids using AI and refer suspected cases to an ophthalmologist or an oculoplastic surgeon. In the future, something like this could be deployed in large What is biomedical informatics? In short, it's the use of biomedical data and technology as tools to enhance human health and well-being. Darvin Yi, PhD, and his team of researchers in the Department of Ophthalmology and Visual Sciences' Artificial Intelligence in Ophthalmology Program are working at the intersection of data science and medical science, leveraging AI and informatics to inform research, fuel innovation and advance patient care. Biomedical Informatics in Ophthalmology Link Bench to Bedside

• Cover