How AI could supercharge your glucose monitor - and catch other health issues

At Stanford, researchers are harnessing the power of artificial intelligence (AI) to enhance our understanding of diabetes, particularly in diagnosing this widespread condition. This could pave the way for more personalized and accessible healthcare solutions.

Diabetes is typically categorized into Type 1 and Type 2, but recent discoveries have revealed crucial subtypes within Type 2 diabetes, which accounts for 95% of cases. These subtypes help predict the risk of developing associated conditions such as kidney, heart, or liver diseases.

Tracey McLaughlin, MD, an endocrinology professor at Stanford, pointed out the challenges in understanding diabetes physiology. "Understanding the physiology behind [diabetes] requires metabolic tests done in a research setting, but the tests are cumbersome and expensive and not practical for use in the clinic," she explained.

By analyzing data from glucose monitors, Stanford researchers have developed an algorithm capable of identifying three out of the four most common Type 2 diabetes subtypes. This algorithm outperforms traditional metabolic tests, accurately predicting metabolic subtypes like insulin resistance and beta-cell deficiency about 90% of the time.

Identifying a patient's specific subtype can significantly influence the effectiveness of treatment. It allows doctors to tailor medication plans and allocate resources more efficiently, potentially lowering healthcare costs. The beauty of this approach lies in its use of data from glucose monitors, which patients often already use, eliminating the need for more complex clinical setups.

"This matters, because depending on what type you have, some drugs will work better than others," McLaughlin emphasized. "Our goal was to find a more accessible, on-demand way for people to understand and improve their health."

This innovation could be a game-changer for the nearly 13% of the US population living with diabetes. By using AI to extract valuable insights from data collected by wearables, this method could dramatically improve treatment options and outcomes, especially for those without easy access to healthcare due to geographical, financial, or other constraints.

Following the spotlight on over-the-counter glucose monitors at CES 2025, where two such devices were recognized as Honorees in Digital Health, this study represents a significant leap forward in making health technology more accessible to everyone.