The eyes are said to be the “gateway to the soul,” but David Kerr, Maryland believes they can also be the gateway to the brain, heart, kidneys and many other organs.
Dr. Kerr is a senior investigator at Sutter's Center for Health Systems Research, studying the role of digital health in diabetes care.
Last year, Sutter Health completed a 6-site, three-month pilot project, conveniently screening diabetic patients with diabetic retinopathy during regular scheduled primary care visits using cameras enhanced by Augmented Intelligence (AI) (often artificial intelligence).
Sutter Health is part of the AMA Health System Member program and provides enterprise solutions that provide leadership, physicians and care teams with resources to promote the future of medicine.
The AI tool implemented by Sutter Health was able to detect signs of retinopathy in less than a minute, allowing patients to undergo screening without dilation of the pupil, allowing them to leave immediately after the test, rather than having to wait more than four hours for the eyes to return to normal. The tool is approved by the US Food and Drug Administration. Across the US, the rate of retinal screening between adults and children is optimal, putting these individuals at risk of losing their vision. The idea behind the project was to make it easier for people living with diabetes to check their eyes each year, as recommended by the American Diabetes Association.
Of the 216 trials conducted, 24 patients showed some degree of retinopathy, 15 listed the condition as “vision threat” and nine were listed as having “mild or higher” problems.
Everything was referred to an ophthalmologist for follow-up.
Sutter Health has since deployed its cameras to 28 primary care offices, and Dr. Kerr believes that the technology can be used to detect other conditions.
“Ai-derived retinal images provide insight into other organs that may be at risk for diabetic complications, such as the heart and kidneys,” he said. “We provide images of blood vessels and nerves. This may be where we can detect diseases much faster before they affect the heart, kidneys and brain.”
Dr. Kerr and his colleagues explained this in detail in an article published last November. Journal of Diabetes Science and Technology.
AI is “the foundation of emerging medical technology. It promotes the future of diagnosing diabetes complications,” writes Dr. Kerr and co-authors.
According to the AMA research (PDF), medical AI uses for specific tasks almost doubled in 2024, and even if questions remain, there is growing enthusiasm for technology.
How AI reduces the risk of complications in people living with diabetes
How AI reduces the risk of complications in people living with diabetes
“Diabetes is a very serious and common chronic metabolic disease and is associated with very poor long-term complications,” Dr. Kerr said.
Dr. Kerr believes that using AI can predict who is at the greatest risk from developing diabetes in the first place, and determines who is at the greatest risk of developing complications for those living with the disease. Evaluating risks early in the natural history of disease provides an opportunity to generate acceptable and cost-effective, evidence-based interventions to significantly reduce these risks. These interventions are increasingly focused on promoting behavioral change and increasing individual self-efficacy.
“What we're seeing is that exponential growth in AI application opens up new ways to help people living with diabetes,” Dr. Kerr said. “That's one of the many potential benefits from AI,” he said.
“We are the beginning of a revolution in diabetes care based on artificial intelligence,” Dr. Kerr said. “AI can see the population, but decide on interventions that reflect the uniqueness of the individual. This is very exciting.”
He acknowledges that “most diabetes treatments currently revolve around prescribing medicines,” but Dr. Kerr said AI could potentially maximize return on investment from non-drug interventions. In the near future, AI will provide more personalized guidance, such as what to choose for breakfast, and the best time to go for that walk is when it helps maintain blood glucose levels within target range.
“From a drug perspective, we can predict whether this particular drug or class of drugs is more effective or less effective in individuals,” he added. “And what about the drug combinations? Do you get even better results by applying the drugs together with plus drug b?”
AMA supports the development of high-quality, clinically validated AIs deployed in a responsible, ethical, transparent way that concerns patient safety above all else.
The revolution in diabetes care
The revolution in diabetes care
“If you stand up over the past 20 years and look at the population, have you managed to reduce the rate of diabetes? No, you haven't,” he said. “So there's no point in continuing to do the same thing.”
While hoping for bold change, Dr. Kerr also noted that SutterHealth has taken a deliberate and responsible approach to AI. Each use of AI is prioritized on patient safety and is carefully vetted to protect privacy, ensure accuracy and maintain transparency, especially when AI plays a role in direct care.
Another important question that doctors have is that AI can fit into my workflow.
Dr. Kerr said the pilot at Sutter Health puts the Ai-Enhanced camera in the primary care office to show that it can be done.
“In diabetic eye diseases, if not diagnosed, can lead to loss of vision and blindness,” he said. “If we wait for a patient to report symptoms, it means that the disease has progressed normally. All we need to do is look for the disease.”
The technology has been used on 28 sites and continues to grow. Now, Dr. Kerr said the key is to use the information the screening provides.
This means closing the loop to ensure that patients who detect screening reach appointments, appointments with specialists and receive timely and appropriate treatment.
The American Academy of Ophthalmology says that screening alone does not improve patient outcomes, and that assessments of the impact of screening programs should also include the number of people referred and examined by ophthalmologists.
“That's the next step in this exercise,” Dr. Kerr said. “The idea is to expand the entire Sutter Health, which has almost 1% of America under its care, 3.5 million patients.”
Dr. Kerr's research highlights the possibility that AI can transform specific areas of care, but it is part of a major effort to think about scaling these technologies across the system, and is now led by Ashley Beecy of Ashleyist, an MD physician who joined Sutter Health at the end of May. In this role, she will oversee the health system's strategies to promote innovation through AI and advanced analytics, helping to improve patient outcomes, support clinicians, and enhance operations.
Learn about the new landscape of healthcare AI. We will also explore how to apply AI to convert healthcare with “Ama Changeded.”® Artificial intelligence from the Healthcare series. ”
