June 2018
University of Iowa Health Care patients are the first in the nation to have access to a new technology that uses artificial intelligence (AI) to diagnose diabetic retinopathy.
The technology, known as IDx-DR, is the only medical device authorized by the Food and Drug Administration that uses AI for the autonomous detection of diabetic retinopathy. The device’s algorithm makes the diagnosis based on imaging of the patient’s retina without the need for an eye specialist to interpret the results.
The technology was created by the Iowa City-based company IDx, a privately held AI diagnostics company, which was founded by Michael Abràmoff, MD, PhD, Professor of Ophthalmology & Visional Sciences and member of the FOEDRC, in 2010. Much of the research used to develop IDx-DR was conducted by Abràmoff at the UI and licensed through the UI Research Foundation.
Two months ago, the company completed clinical trials with 900 patients in hopes of receiving FDA approval early next year. If approved, Abràmoff said, the tool could preserve the vision of countless diabetes patients, save millions in health care costs, equip primary care providers with a relatively easy-to-use tool and pave the way for technology like it.
“Screening for diabetic eye complications (diabetic retinopathy) is an essential component of comprehensive diabetes care. “Innovative strategies such as autonomous diagnostic AI represents an important advance,” adds E. Dale Abel, MD, PhD, UI professor and DEO of Internal Medicine, director of the Division of Endocrinology and Metabolism with UI Health Care, and director of the FOEDRC. “Importantly, this will allow patients to be screened for diabetic retinopathy at the same time as their regular visit with their provider. Moreover, it will identify those patients with significant retinopathy who need to be referred to eye specialists in a timely manner and determine those who do not. This innovation further strengthens our ability to provide state-of-the-art care for our patients with diabetes.”
People with Type 1, Type 2, and gestational diabetes are at risk for diabetic retinopathy, which affects 7.7 million Americans and is a leading cause of blindness in the U.S. Early detection and treatment can reduce the risk of blindness by 95 percent, but diabetic retinopathy often lacks early symptoms, so regular screening is needed to diagnosis the condition.
Previously, screening would require an appointment with an eye specialist. The new AI-based system can be used by trained primary care providers during routine office visits, with results available in minutes. If a patient is diagnosed with more than mild diabetic retinopathy, they are referred to an eye specialist for follow-up and treatment.
It is important to note that the IDx-DR system only detects diabetic retinopathy; it is not designed to diagnose other forms of diabetic eye disease and not intended to replace a comprehensive eye exam with an eye care professional.
However, it provides a convenient option for people with diabetes to complete their recommended annual diabetic retinopathy exam. This is particularly important for the more than 50 percent of people with diabetes who are at high risk for vision loss and do not see their eye doctor on a yearly basis.