Michael Abramoff, MD, IDx founder and president, mentioned in a statement that, “The AI system’s plays a vital role to identify those people with diabetes who are similar to have diabetic treatment requires further evaluation by an eye-care provider for retinopathy” and this study also demonstrate the safety of autonomous AI systems to provide advanced diagnostics to a primary care clinics with an potential to reduce cost and increase fast access.
Dale Abel, MD, director of the division of endocrinology and metabolism at UI Health Care mentioned in a statement that “Diabetic retinopathy in the early stages can be identified in an essential component of comprehensive diabetes care that strengthens our ability to provide patients with diabetes at different stages.”
IDx-DR’s algorithm takes the images with the Topcon NW400 retinal camera and uploaded to a cloud server within minutes, the software provides doctors a binary report that indicates more than mild diabetic retinopathy and that the patient should be referred to an eye care specialist, or that the screen is negative.
The team collected the retinal images of 900 diabetes patients from 10 U.S. sites which were taken by AI-assisted primary camera that were installed into IDx-DR’s algorithm for diagnosis. “Early detection of retinopathy is very important for people with diabetes to treat instantly without giving any chance for diabetes to develop further. “Today’s decision permits Artificial intelligence technology that can be used in a primary care doctor’s clinic and FDA will continue to facilitate its availability of safe and efficient digital health devices that may improve patient personal health.”
He also mentioned that 40 participants excluded from the analysis could have a significant impact on their specific diabetic study that will make them wonder how the tool’s sole capacity for diabetic neuropathy diagnosis will work in the realistic scenarios where patients could present with a number of other retinal problems.
The new-approved device will have significant uptake in the clinic that is expensive and it remains to be determined how and where it should be implemented,” and they also stated that FDA’s De Novo premarket review pathway, by necessity, allows a bit more leeway than the standard 510(k) clearance. Investigations are likely to develop more in prior to present Artificial Intelligence in the field of diagnostics
As a result, it will be incumbent on healthcare personals to become more attracted with this and other AI technologies that hit in the upcoming years to ensure that they are used for early detection of diabetes and hope this study strives a path in regards to that study