Diabetes is a horrendous disease. Most people associate it with changes in blood sugar and obesity. However, it also causes an array of long-term conditions. Diabetic retinopathy is one of them. It occurs when the tiny blood vessels in the eye are damaged by consistently elevated blood sugar levels.
Every year, 60,000 people go blind because of diabetic retinopathy. A startup called Digital Diagnostics is working to help those patients retain their vision by using the power of artificial intelligence (AI).
Michael Abramoff, a retinal surgeon and the founder of Digital Diagnostics, set out to revolutionize the way that diabetic retinopathy is diagnosed. During his training, he found that patients with the disease often went undiagnosed until it was too late for effective treatment. Ten years ago, he started Digital Diagnostics.
The startup was originally called IDx since its product is called IDx-DR—the DR stands for diabetic retinopathy. Essentially, the tool takes images of the back of the eye, analyzes them, and provides a reliable diagnosis within minutes. If the system detects a moderate or severe case, it automatically refers the patient to a specialist.
IDx-DR uses Nvidia GPUs and the company’s deep learning tools to accelerate the image analysis process. The approach allows Digital Diagnostics to perform testing in easily accessible settings like primary care offices and drugstores.
John Bertrand, the startup’s CEO, says, “Moving care to locations the patient is already visiting improves access and avoids extra visits that overwhelm specialty physician schedules. Patients avoid an extra copay and don’t have to take time off work for a second appointment.”
IDx-DR is the first FDA-cleared autonomous AI system. Although the agency has approved many non-autonomous AI solutions, IDx-DR is different because it doesn’t require physician oversight.
“There are lots of good AI products specifically created to assist physicians and increase the detection rate of finding an abnormality but to allow physicians to practice to the top of their license, and reduce the costs of these low complexity tests, you need to use autonomous AI,” Bertrand says.
By improving access to diabetic retinopathy testing, Digital Diagnostics’ system helps prevent the disease from progressing past the point of treatment. In the end, patients may be able to retain their sight by catching the disease early.
Already in Action
Digital Diagnostics’ AI-powered system is already making an impact in the real world. The platform has been deployed in more than 20 sites across the U.S.
The Blessing Health System in Illinois is one network using IDx-DR. Dr. Tim Beth says, “Digital Diagnostics has done well in developing an algorithm that can detect the possibility of early disease. We would be missing patients if we didn’t use IDx-DR.”
Moving forward, Digital Diagnostics will look to apply the same AI-based approach to other conditions. The startup is working on prototypes for diagnosing glaucoma and age-related macular degeneration. It is also hoping to expand past the eye and develop solutions for diagnosing skin conditions and diseases affecting the nose and throat.