February 2018

Diabetic retinopathy (DR), a complication of diabetes, is the leading cause of preventable blindness in the developed world, and it is one of the most feared complications for people with diabetes. In the US, at least 25,000 people with diabetes go blind every year from this almost entirely preventable disease, and there are 25 million Americans with diabetes at risk for the disease, projected to increase to 50 million over the next 10 years. There is extensive proof that an annual eye exam to detect the retinopathy early and treat it before the onset of symptoms can prevent almost all permanent visual loss, unfortunately the annual eye exam is expensive and access may be difficult.

For the past 20 years, Dr. Michael Abramoff, MD, PhD, who is both a retinal specialist and a computer scientist, in the Department of Ophthalmology at the University of Iowa and a member of the FOE Diabetes Research Center, has studied how Artificial Intelligence (AI) can be used to prevent diabetic retinopathy and thus improve the lives of people with diabetes.

One of his research programs studies how smart cameras can take pictures of the inner lining of the back of the eye, the retina, and use AI detect whether or not there is diabetic retinopathy. His early studies showed that AI computer algorithms can do this better than retinal specialists do. Other studies showed how such cameras can be made at lower cost, so that they are affordable for the frontlines of care such as primary care. Thus, these smart cameras could be placed in the offices of family doctors and primary care providers. His research led him to patent the resulting computer algorithms and cameras, and eventually, to found a company, IDx, to bring this technology to benefit people with diabetes, making prevention more accessible at increased quality and lower cost.

IDx-DR has not yet been cleared by FDA. However, this smart camera system has the potential to truly bend the healthcare delivery cost curve by bringing diabetic retinopathy screening to the frontlines of medical care. It is hoped that then only those people with diabetes who have diabetic retinopathy will be referred to an eye care specialist for further treatment and management. The IDx-DR clinical trial was completed several months ago, and FDA clearance may come as soon as spring 2018 – it already cleared and in use in Europe.

Another research program studies the relationship between the damage to the small blood vessels in the retina, that is characteristic for Diabetic Retinopathy, and the damage to the nerve cells in the retina, that transmit visual information to the brain, a disease called retinal diabetic neuropathy (RDN). Using novel and innovative imaging strategies Dr. Abramoff’s team has demonstrated how this process occurs and the time frame over which this occurs. There is currently no known prevention or treatment for RDN, in part because its significance in diabetic eye disease was not widely appreciated. Recently they found intriguing evidence in people with diabetes, as well as in mice with diabetes, that in the eye and the brain, damage of the nerve cells occurs before there is any damage to the small blood vessels. Currently he and his team are studying whether this nerve damage can be prevented, and if prevention of the nerve damage can prevent diabetic retinopathy. Dr. Abramoff's research has been cited over 22,000 times. In October 2017, he was asked to brief members of the US Senate and House in DC on the critical role that basic science funding plays in the development of new healthcare innovations, and on the role that AI can play in driving down healthcare costs.