The FDA has, for the first time, approved an Artificial Intelligence diagnostic device, that can detect diabetic retinopathy without the involvement of an eye doctor. This software programme can diagnose diabetic retinopathy by scanning photographs of the retina. The programme is called IDx-DR and was developed by Michael Abràmoff, MD, PhD. The FDA approved it after a trial of nine-hundred patients, across 10 primary care sites confirmed the accuracy of the results. The IDx-DR correctly detected retinopathy about eighty percent of the time, and could correctly identify those who didn’t have the disease about ninety percent of the time. Fundus photos are up-loaded, and provided the quality is good enough, the software algorithm will make the diagnosis.
Michael Abràmoff said:
“It makes the clinical decision on its own. This means that the technology can be used by a nurse or doctor who’s not an eye specialist, making diagnosis more accessible. For example, patients wouldn’t need to wait for an eye specialist to be available to get a diagnosis, ” said Abramoff. “This represents one component of the work we are doing to develop an autonomous AI diagnostic system that accurately identifies patients at high risk of cardiovascular events.”
IDx-DR is part of a growing trend of algorithms learning how to spot and diagnose disease. Measurement of arteries and veins have long been studied as important predictors of an increased risk of stroke, cerebral atrophy, cognitive decline, and myocardial infarction. Yet, these measurements, when done by human experts are often inaccurate and may result in missing early warning signs that would allow physicians to initiate outcome-changing interventions. Using AI, the new patented method and systems overcomes these limitations to obtain more accurate measurements.