The University of Virginia (UVA) is set to begin a groundbreaking clinical trial testing an artificial intelligence-powered device designed to revolutionize Type 1 diabetes management. The FDA-approved study, scheduled to commence in March, will evaluate a novel reinforcement learning feature called the "Bolus Priming System with Reinforcement Learning" (BPS_RL).
Advanced Technology Integration
The BPS_RL technology seamlessly integrates with the existing Automated Insulin Delivery Adaptive NETwork (AIDANET), which combines a smartphone application, Dexcom continuous glucose monitor, and Tandem insulin pump. What sets this system apart is its ability to deliver insulin automatically without requiring user intervention, potentially addressing one of the most significant challenges in diabetes management.
Dr. Heman Shakeri, Assistant Professor of Data Science at UVA, emphasizes the transformative potential of this technology: "We are committed to creating a fully automated, intelligent insulin delivery system that redefines diabetes management, making treatment simpler, more reliable, and entirely effortless for patients."
Trial Design and Methodology
The study employs a comprehensive three-week protocol involving 16 adult participants who have prior experience with automated insulin delivery systems. The trial structure includes:
- Week 1: Home-based baseline assessment using the standard AIDANET system
- Week 2: Supervised testing location sessions comparing standard and enhanced systems (18-hour periods each)
- Week 3: At-home testing of the enhanced system under remote monitoring
The trial utilizes a crossover design, with participants randomly assigned to begin with either the current or new system before switching to the alternative.
Addressing Unmet Needs
For individuals living with Type 1 diabetes, maintaining optimal blood glucose levels presents constant challenges. Insulin requirements fluctuate based on multiple factors, including meals, physical activity, stress levels, and other variables, making precise dosing particularly challenging. Current automated insulin delivery systems often require significant user input and can be both expensive and difficult to access.
Technical Development and Expertise
The trial brings together an accomplished research team, including Boris Kovatchev, founding director of the UVA Center for Diabetes Technology, and Anas El Fathi, research assistant professor at the Center for Diabetes Technology. Postdoctoral researcher Ali Tavasoli played a crucial role in fine-tuning the BPS_RL system and developing the computer simulations that supported FDA approval.
Future Implications
Beyond immediate glycemic control improvements, the research team aims to address broader challenges in diabetes care. The development of more adaptive, precise, and cost-effective insulin delivery systems could significantly reduce both the mental and financial burden associated with diabetes management. This initiative represents a significant step toward making diabetes care more efficient and accessible for the broader patient population.
