Supported by a 2023 CITRIS Seed Award, UC Davis electrical and computer engineer Samuel King and UC Davis Health pediatric endocrinologist Stephanie Crossen are working to reduce the cognitive burden on individuals managing Type 1 diabetes (T1D) by piloting software that alerts users to anticipated glucose changes before they occur.
T1D is an autoimmune disease that prevents the pancreas from creating its own insulin, a hormone that helps blood glucose enter the body’s cells to be used as energy. This requires people with T1D to constantly monitor their blood glucose levels, often with wearable continuous glucose monitors (CGMs), so they can inject themselves with the insulin their bodies are unable to make.
King, who was recently diagnosed with T1D himself, found that his CGM was only able to alert him once he was already in glycemic imbalance — an urgent situation that required immediate action.
He and his research team were inspired to develop a proactive “metabolic watchdog” that they call BeaGL. Designed to work with CGMs, the BeaGL software employs security-focused machine learning algorithms to anticipate glucose changes and send the user predictive alerts via a device, such as a smartwatch.
BeaGL particularly seeks to assist young adults, the age group where glucose management tends to suffer most. Six UC Davis students with T1D have been using BeaGL in the research phase with beneficial results, enabling them to build trust in the system and allowing King to integrate user feedback.
King and Crossen see BeaGL as a promising first step toward a closed-loop insulin delivery driven by AI and hope BeaGL can inform a broader integration of AI into health care.