Social support and machine learning are at the core of a student-developed app for people with diabetes

Diabetes is the seventh leading cause of death in the U.S., and nearly 10% of the population suffers from this chronic disease, according to the Centers for Disease Control and Prevention. Yet despite the disease’s prevalence and potentially deadly consequences, studies show that nearly 50% of people with diabetes don’t adhere to recommended self-care and treatment regimens.

Harvard students have set out to change that by creating a social platform to motivate people with diabetes to keep up with glucose monitoring, diet, exercise, and other healthy habits.

Their startup, Checkmate Diabetes, grew out of the Harvard College course Glucose: from Molecules to Society (SCRB 175), explained co-founder Emi Gonzalez, A.B. ’18, a molecular and cellular biology concentrator. After hearing from amputees and individuals who lost their vision due to poor diabetes management, the class embarked on an open-ended project to find a solution.

So Gonzalez gathered a team and applied to the Harvard Innovation Labs to turn their idea into a reality. They are developing a mobile app that uses gamification and community-building to motivate people with diabetes, explained Filip Michalsky, M.E. ’19, a student in the computational science and engineering master’s program offered by the Institute for Applied Computational Science (IACS) at the Harvard John A. Paulson School of Engineering and Applied Sciences (SEAS).

In the current beta version, Checkmate users track blood sugar levels and receive an abstracted score based mostly on how often they check. Seeing their ranking compared to others’ scores is a motivating force, Michalsky said.

In addition to the competitive aspect of the app, Checkmate will offer tangible rewards for users with the highest scores; prizes will include diabetes test strips, test kits, and gym passes.

The app also builds community. Checkmate’s location-based system connects users to other people with diabetes who live nearby, or are in the same age group, encouraging the formation of support groups.

The next version will include a more robust score, incorporating how well a user adheres to dietary restrictions and an exercise regimen. Those additions will help Checkmate stay relevant in the future, as more patients begin using blood glucose monitoring systems.

The development team is also incorporating machine learning, enabling the app to learn what information is most likely to motivate a specific user. Maybe seeing higher scores inspires someone to strive harder, or perhaps seeing users with similar scores, who face relatable struggles, is more motivating, Michalsky said.

Harvard John A. Paulson School of Engineering and Applied Sciences has the full story