Development and optimization of HEDRA, a novel m-health decision support tool for weight and glycemic control in Type 1 Diabetes

Michael Kosorok, Eric Laber, Beth Mayer-Davis and Dr. Laura Young, who specializes in the care of patients with diabetes, were recently awarded funding for this project, which aims to develop a mobile health app for patients with Type 1 diabetes.

The app will help these patients "more easily engage in physical activity" by using data it collects about them to "provide personalized strategies before, during and after planned physical activity, to keep blood sugar levels in a normal range," said Anna Kahkoska, an M.D./Ph.D candidate at UNC-Chapel Hill, co-mentored by Mayer-Davis and Kosorok, whose dissertation focuses on the use of machine learning techniques to identify Type 1 diabetes patients with distinct disease subtypes who might respond better to alternative treatment options.

Laber, an associate professor in NC State's Statistics Department, explained that the statistical expertise will play a heavy role in determining how and when to implement these strategies, such as sending a reminder to take one's insulin, so as to optimize the app's effectiveness.

"Mobile health has a lot of potential . . . because you can deliver messages when and if they're needed," Laber said. "It's kind of a neat area because you can reach so many people — it's kind of a way of scaling healthcare."

Moreover, this particular project "represents a new direction for a transdisciplinary collaboration" in the way it has brought together researchers who specialize in a wide range of disciplines — from nutrition and endocrinology to biostatistics and software engineering.

However, this unique collaboration also results in a niche kind of research for which funding can be hard to find through "traditional" sources.

That's where the NC TraCS Pilot Grant Program came in.

"We knew we needed proof of concept (for the app)," Laber said. "And it was going to be hard to get (funding for) that any other way."

Laber said that for projects like this — which are "high-risk-high-reward" but also "a little unusual," making traditional funding mechanisms not a "good fit" — he thinks the NC TraCS Institute Pilot Grant program is "perfect."

See also: How Carolina is transforming medicine with artificial intelligence