An Android Phone Based Artificial Pancreas System for Type- 1 Diabetic Patients

An Android smartphone based software architecture for an Artificial Pancreas (AP) system for Type-1 diabetic patients is proposed in this paper. This architecture incorporates an interacting two-application (two-App) approach. Each Application has unique functionalities and is developed through different processes for rapid prototyping. Application A (the front-end App), which acts as the user interface as well as communication hub for the hardware devices, was developed using the Android Studio. The Continuous Glucose Monitor (CGM) sensor communicates to the phone via the Near Field Communication (NFC). The insulin pump, on the other hand, communicates through an intermediate device which converts the Bluetooth signal from the phone to Radio Frequency (RF) signal, which can then be received by the pump. The compatibility libraries to establish this communication is in-built into Application A. Application B (the back-end App) which contains the algorithm modules for computing insulin using the advanced model predictive control philosophy, was generated using Simulink and quickly customized using Android Studio with relative ease, details of which are included in this paper. Various functionalities of this novel two-App system architecture was extensively verified and validated as well from in-Silico experiments.