Eyeworld

72 | EYEWORLD | SUMMER 2025 G UCOMA Contact Murphy: jamestmurphyiiimd@gmail.com Ou: Yvonne.Ou@ucsf.edu Provencher: lorrainemprovencher@gmail.com and are more user-friendly when it comes to connecting to your EHR. Unlike larger, tabletop devices, wearable devices can be mailed in for service, immediately replaced, and often remote WiFi-based troubleshooting can solve problems; software updates come automatically over WiFi. Larger devices cannot be shipped, so when they require service, a service technician has to travel to your office, often taking weeks, which is a big hit to a practice's workflow and bottom line, Dr. Murphy said. From an efficiency standpoint, Dr. Murphy said completing a virtual visual field with a wearable takes between 2–3 minutes. Yvonne Ou, MD, is currently studying virtual reality visual fields but not making any clinical decisions based on them. "Through an originally unfunded collabora- tion and now an NEI-sponsored project, we are working with Vivid Vision to study Vivid Vision Perimetry (VVP) and its ability to accurately monitor glaucoma progression," Dr. Ou said. "I do not think there is any clinical evidence (as of yet) that virtual reality visual fields can be used for monitoring glaucoma progression, and I hope to be part of the team that generates that evidence." Dr. Ou said the potential benefits of virtual reality visual fields are numerous: easier test taking, more efficient use of clinic resources, lower cost, portability, and home testing. In addition, the Vivid Vision test is easier to take for patients as it does not require patients to maintain steady fixation and suppress their foveation reflex. Instead, the test is designed so that patients look at each stimulus as it is being presented, making for a more intuitive test tak- ing experience. Artificial intelligence in glaucoma While none of the physicians interviewed for this article use artificial intelligence directly in the clinic to inform their glaucoma practice currently, they do envision a future where it will be helpful. "I think AI will continue to grow in all areas of medicine, but for glaucoma in particular, gen- erative AI could help clinicians synthesize the immense amount of data we use to make deci- sions (e.g., family history, pachymetry, vision, corneal hysteresis, optic nerve exam, IOP, OCT, visual fields, etc.) to risk stratify patients and support more efficient decision making. It could also monitor for progression in testing, either in person or even remotely," Dr. Provencher said. She added that if AI could someday screen patients and accurately triage them to a primary eyecare provider or a surgeon, this would be a huge win for the aging population and stretched medical system. "Also, we spend a large proportion of our time counseling and educating patients. AI could help with this to make the doctor more efficient, but it must not come at the expense of the patient experience and the personal con- nection a patient needs with their surgeon," Dr. Provencher said. Dr. Ou is excited to see the potential of AI identifying patients at risk for glaucoma, making screening programs more efficient and widespread. "I am also interested in the po- tential of AI to identify and predict high-risk progressors, as these are the patients who need timely intervention," she said. "But diagnosing glaucoma and identifying its progression is not always easy even for the most experienced glau- coma specialist, thus great care must be taken as we adopt AI into clinical practice." Dr. Murphy said there is already an FDA-ap- proved AI-based tool for glaucoma screening, Alris AI. He said this tool, which he doesn't have experience with, crunches numbers like corneal thickness, visual field, retinal nerve fiber layer, and IOP over time to provide a glaucoma risk assessment. Other technologies in the pipeline that will be helpful for glaucoma include handheld/ mobile OCT. The theme here is smaller, faster, cheaper is better, Dr. Murphy said. "AI-based analytic tools have already demonstrated that they are capable of distinguishing glaucomatous versus non-glaucomatous field defects with ac- curacy as good as or better than trained subspe- cialists, and the potential is only growing. It's already smarter than we are, it's just a matter of how practices leverage that potential to their benefit and the benefit of patients," he said. Dr. Murphy said he would love for AI tech- nology to be able to help him compare apples to oranges of 20+ years of visual field analysis from an older proprietary progression analysis system to newer wearable systems. "AI has such a wide-ranging number of ap- plications, and we're just scratching the surface of what it can do for us," Dr. Murphy said. continued from page 71

• Cover