DBLG2

DBLG2 is an Android mobile application for managing type 1 diabetes in persons 12 years and older. It works with compatible continuous glucose monitors and insulin pumps to automatically adjust basal insulin delivery every 5 minutes based on glucose readings and predictions, and can deliver correction boluses. The system also includes a meal bolus calculator and self-learning algorithm to optimize insulin dosing.

DBLG2 uses a hybrid closed-loop predictive control algorithm that predicts glucose levels up to 5 hours in the future (versus Tidepool Loop's 6 hours) and insulin delivery up to 2 hours ahead. It operates on Android platforms whereas the predicate uses iPhone. DBLG2 uses a glucose target set point with customizable thresholds rather than a target range, includes automatic target adjustment during physical activities, and has a self-learning module for long-term parameter optimization.

IEC 62304:2006+A1:2015 (software lifecycle), IEC 62366-1:2015 (usability engineering), ISO 14971:2019+A11:2021 (risk management), ISO 14155:2020 (clinical investigation), IEC 81001-5-1:2021 (health IT safety), EN ISO 20417:2021 (labeling), and FDA guidance on software validation and cybersecurity management.

DBLG2 is substantially equivalent to Tidepool Loop because both are hybrid closed-loop interoperable automated glycemic controllers for type 1 diabetes using the same regulatory classification and product code. They share the same intended use, algorithm type, compatible iCGM (Dexcom G6), and functional safety features. Design differences (Android vs. iPhone, modified prediction windows, target approach) do not raise new safety/effectiveness questions. Clinical data from 16,919 patient-weeks supports safety and effectiveness comparable to the predicate's 4,550 patient-weeks.