Type 2 diabetes varies greatly among individuals, but now, researchers at Stanford Medicine have developed an AI-based tool that uses data from continuous blood glucose monitors to identify different subtypes of this condition. This breakthrough could lead to more personalized treatments and better health outcomes.

Dr. Michael Snyder, a professor of genetics, explained, “It’s a tool that people can use to take preventative measures. If the levels trigger a prediabetes warning, for instance, dietary or exercise habits could be adjusted.”

Approximately 13% of the U.S. population has diabetes, and 98 million have prediabetes. This new AI technology could be a game changer for diabetes care, offering detailed diagnostic information that can lead to personalized treatment plans.

Dr. Tracey McLaughlin, a professor of endocrinology, said, “The majority of people with diabetes have Type 2, but it’s more complex than that, and there are different underlying causes. Our goal was to find a more accessible, on-demand way for people to understand and improve their health.”

Using a continuous glucose monitor, a standard device worn on the upper arm, the AI algorithm analyzes blood sugar patterns to predict different subtypes of Type 2 diabetes. In a study involving 54 participants, the AI tool accurately identified subtypes such as insulin resistance and beta-cell deficiency about 90% of the time.

This new technology not only provides higher-resolution data for diabetes care but can also help people with insulin resistance, which is a risk factor for other health conditions like heart disease. The researchers hope that this widely accessible technology will improve health care for those who are economically challenged or live in remote areas.