AI may play role in point-of-care screening for diabetic retinopathy
Click Here to Manage Email Alerts
KOLOA, Hawaii — Point-of-care screening tools may help improve screening compliance for diabetic retinopathy, according to a speaker here.
“We recognize that there is a global epidemic of diabetes. When you look at the rates of projected diabetics, it’s pretty staggering,” SriniVas Sadda, MD, said at Retina 2023. “We will have to develop systems to cope with this.”
Sadda said there were 415 million people with diabetes in 2015, which is expected to grow to 642 million people by 2040. Approximately 90% of diabetic blindness can be prevented with early detection and treatment.
Annual screening is critical for these prevention efforts, Sadda said. Smartphones are one potential tool for screening.
“It’s great that we’ve gone to herculean extraordinary measures to try to capture the imaging information, but still there is the issue of actually getting it to a place for interpretation,” he said. “Many of these models rely on reading centers, which of course can potentially add an additional source of delay and expense.”
Going forward, autonomous AI-based image analysis may be a helpful component for screening measures, and AI-based solutions may offer the ability to generalize to multiple imaging systems.
“I do believe that these autonomous AI-based systems are going to be a key component,” Sadda said. “We all need to think about how we are going to incorporate them into our practice models.”