Detecting failures of the glucose sensor-insulin pump system: improved overnight safety monitoring for Type-1 diabetes.

New sensors for real-time continuous glucose monitoring (CGM) and pumps for continuous subcutaneous insulin infusion (CSII), possibly mounted on the same device, opened new scenarios for Type-1 diabetes treatment. However, possible failures of either CGM or CSII can expose diabetic patients to risks that can be dangerous especially overnight. In this contribution we present a proof-of-concept method, developed in a state-space context and implemented through a Kalman estimator, to detect in real time possible overnight failures of the sensor-pump system by simultaneously using CGM and CSII data. The method is tested on two simulated and one real subject. Results show that the method is able to correctly generate alerts for sensor-pump failures and stimulates further investigation on its development.