The Center for Medicare and Medicaid Services (CMS) will cover an autonomous screening tool and allow it to be included in quality measures.As Part of its 2022 Medicare Physician Fee Schedule, CMS has set a national payment amount for the use of AI to detect diabetic retinopathy.
It also includes these screenings as a quality measure, stating that eye examinations must be performed by an ophthalmologist or optometrist, or that retinal images must be read by a system that provides AI interpretation.
Diabetic retinopathy is a condition in which high blood sugar levels cause damage to blood vessels in the retina. It can go undetected until later, when it starts to affect the patient’s vision.
because of this, It is recommended to have an eye exam every year Suitable for patients with type 1 or type 2 diabetes. But in practice, this does not always happen. A study of hospitalized diabetic patients found that Less than half them I had an eye exam last year.
Some commentators argue that quality measures should only be applied when the software system is approved by the FDA, or that comprehensive eye examinations and artificial intelligence-based screenings are considered to indicate that the quality of care is inappropriate.
In response, CMS wrote: “We believe that allowing AI will increase accessibility, thereby increasing the incidence of annual eye exams, and thereby bringing high-quality results to this patient population.”
The decision was also cheered by two companies, which have developed artificial intelligence-based screening systems Digital Diagnostics and Eyenuk for diabetic retinopathy.
“These final regulations will allow eye examinations at the point of care of patients, where they receive diabetes management, thereby removing many barriers to care and increasing accessibility,” said Michael D Abramoff, PhD, founder and chairman of Digital Diagnostics Said in the press release.
Digital diagnosis (formerly known as IDx) Obtained de novo license from the U.S. Food and Drug Administration in 2018 Because its software can automatically detect diabetic retinopathy. Images taken with a retina camera are uploaded and processed by its algorithm, which screens for moderate to severe retinopathy. This will indicate to the doctor that there is diabetic retinopathy or the screening result is negative.
It is worth noting that the device has been approved to make screening decisions on its own, which means that primary care physicians and other providers who do not specialize in eye care can use it.
Although the doctor In nature editorial Touted customs clearance as an important milestone, they also pointed out some shortcomings of these automated tools. For example, they write, the research that is the basis of the FDA’s decision for the device is still relatively small in terms of diagnostic accuracy, and because it is only used to detect diabetic retinopathy, it may miss ophthalmic examinations for other conditions, such as glaucoma or macular degeneration.
As far as Eyenuk is concerned, Obtained 510(k) license Its system last year.
Photo credit: memorisz, Getty Images
