KIAWAH ISLAND, S.C. — In this Healio Video Perspective from Kiawah Eye, Joe Luvisi, MD, discusses the use of AI to predict response to anti-VEGF therapy in diabetic macular edema.
Luvisi and colleagues used a convolutional neural network to find out if they could predict which treatment-naive patients would not respond to anti-VEGF therapy after their first treatment. The AI model predicted 55 out of 57 patients who responded and eight out of 16 patients who did not.
“It proved a useful tool in trying to find patients that are going to respond to anti-VEGF on their initial treatment vs. patients that are not,” he said. “It’s early on in our integration with using AI models in medicine. As we do further studies and analysis of this type of stuff, we hope that this is at one point something we can utilize in our clinic, whereas a clinician I can make a more informed decision on what patient and what treatment and what’s going to be best for them.”