Emerging research is showing how artificial intelligence (AI) and machine learning applications can be used to analyze the continuous glucose monitoring (CGM) data of diabetes patients and help clinicians deliver more personalized care and lifestyle recommendations.

A Scientific Sessions panel discussion will include a pair of experts who will discuss the latest advances in technology and the future of CGM. From Glucose Streams to Phenotype Dreams—The Future of Continuous Monitoring will be held on Sunday, June 22, from 4:30–6:00 p.m., in Room W196 BC of the McCormick Place Convention Center.

Michael Snyder, PhD, the Stanford W. Ascherman Professor of Genetics and Director of the Center for Genomics and Personalized Medicine at Stanford University School of Medicine, will discuss how CGM data is transforming diabetes subtyping and management.

“We believe there are many subtypes of diabetes, especially type 2. There is muscle insulin resistance, there are beta cell defects, incretin defects, and hepatic insulin resistance, just to name a few examples,” Dr. Snyder said. “For the most part, however, when it comes to diabetes management and care, it all has been lumped together simply as type 1 or type 2.”

As part of their ongoing research, Dr. Snyder and his colleagues have developed an AI-based algorithm that can identify varying subtypes of type 2 diabetes by analyzing CGM data.

“Continuous glucose monitors are very powerful, and if you look at people’s curves, they have very distinct shapes, and we believe those shapes reflect the underlying biology of the glucose regulation defects that people have,” he said. “We are finding, using machine learning, that we can actually subtype people just from their curve with pretty good accuracy. We can identify muscle insulin resistance, and to some extent, we can identify liver insulin resistance and incretin effects.”

Using that knowledge, Dr. Snyder said, can help clinicians deliver more precise, personalized care, including prescribing medicines likely to be the most effective, as well as providing individually tailored lifestyle and dietary recommendations.

“For example, as part of our study, we had people in seven different groups eat seven different carbohydrates and then measured the extent to which their blood sugar spiked,” he said. “Of course, it’s long been known that people will spike to different foods, but it has not really been demonstrated in so well-controlled a way as we did in our study. What we discovered is that not only do different people spike to different foods, but we could actually correlate it with a distinct subtype.”

That correlation exemplifies the heterogeneity of diabetes and highlights the importance of individualized care.

“We are going to become increasingly able to start telling people, based on their subtype, what foods they should and shouldn’t eat and what lifestyle changes or choices might benefit them most,” Dr. Snyder said. “We’re going to know exactly what’s going on with people and then be able to make very specific recommendations about what they might do about it. That is the ultimate in precision diabetes.”

Also during the session, Celeste Durnwald, MD, Associate Professor of Obstetrics and Gynecology at the Hospital of the University of Pennsylvania and the University of Pennsylvania Perelman School of Medicine, will discuss novel uses of CGM in gestation and gestational diabetes. After the presentations, Dr. Durnwald and Dr. Snyder will join session chair Rena Belfort De Aguiar, MD, PhD, Associate Professor, University of Texas Health San Antonio, for a panel discussion.

On-demand access to recorded presentations will be available to registered participants following the conclusion of the 85^th Scientific Sessions, from June 25–August 25.