IoT technologies combining glucose control with physiological signal: comparative study

Abstract

Patients with type 1 diabetes mellitus (T1DM) have varying sensitivities to insulin and also varying responses to meals and exercises, an Artificial pancreas (AP) which is a closed loop system are used to control blood glucose concentration. With advances in continuous glucose monitoring (CGM) technologies, intelligent control and communication systems, AP have improved better postprandial glucose. Despite these advances, many researchers have developed a system able to keep Blood glucose concentration (BGC) in the target range during all the different situations (stress, during and after exercise and overnight$\dots$ etc). These different situations present a major challenge in the development of closed loop AP system, because of their effect on the BGC are not well understood. IoT emergent technologies allow to create new trend in the AP system introducing physiological signals to the closed loop system. This started with the few studies that found some correlation between physiological signals such as electrocardiography (ECG), electroencephalography (EEG) and changes in BGC during different situations. many researchers aim to develop an Intelligent control system that predict and avoid automatically hypoglycemia and hyperglycemia episodes using biometric variables extracted from the physiological signal instead of CGM. In this paper we will present an overview and a comparison study between the different studies that use these physiological signals in AP systems, concluding that the ECG signal are the most appropriate physiological signal that can be used in combination with glucose control strategy for better prediction and prevention of hypoglycemia and hyperglycemia episodes.