Artificial intelligence begins diagnosing patients in Eastern Iowa
First of its kind device created by IDx to detect diabetic retinopathy
CORALVILLE — The University of Iowa Hospitals and Clinics has become the first to employ new technology — developed by a company rooted in the university’s research engine — that uses artificial intelligence to diagnose an eye disease.
The device, which received approval from U.S. Food and Drug Administration in April, was developed to diagnose diabetic retinopathy, a diabetes complication that can cause blindness.
Called IDx-DR, the device uses software and a retinal camera to take images of a patient’s retina. From there, the AI analyzes the patient’s images “the same way I do as a clinician” to determine if the patient has the condition, said Dr. Michael Abramoff, president and director of IDx and UI Health Care ophthalmologist.
“It looks for different lesions like hemorrhages, microaneurysms, many other abnormalities you get from diabetes in the retina if it’s abnormal, which is what I do when I look for a patient,” he said.
“Then it analyzes the combination of all these different features and it gives you a clinical decision by itself.”
The Diabetes and Endocrinology Center has approximately 7,200 patient visits per year, according to UI officials in a news release on the device.
Diabetic retinopathy can affect anyone with Type 1 or Type 2 diabetes, and is one of the leading causes of blindness in the United States, according to the federal Centers for Disease Control and Prevention.
Early detection is key, Abramoff said. However, as diabetic retinopathy lacks early symptoms, individuals must undergo regular retinal exams to spot the condition.
Abramoff hopes his device can make that process easier by providing accessibility to this exam outside of an eye specialist. IDx-DR is used by provider staff during routine visits and can provide results within moments.
Then, if IDx-DR detects more than mild diabetic retinopathy, the patient is referred for follow-up examinations and treatment with an eye specialist.
According to his clinical trials with 900 patients before the FDA approval, Abramoff said the device had an 87 percent sensitivity to detecting the disease in all types of patients.
Any procedures or treatments to be conducted by an autonomous AI are far off, but Abramoff said he is hopeful the technology will expand to more clinics and other health care settings.
This, he believes, could be the future of detecting and diagnosing other diseases.
UI Health Care plans to expand use of this device across the system, university officials said in the news release.
“I see a great future,” Abramoff said.