Physiological Models and Control for Type 1 Diabetes Mellitus: A Brief Review

摘要

The regulation of blood glucose in Type-1 Diabetes Mellitus (T1DM) patient is being extensively investigated by researchers. This research, therefore, has contributed to the development of many glucose-insulin mathematical models, which at some level successfully mimic the physiological behaviour of the human body. These mathematical models describe the dynamics of glucose in the presence of insulin in the body. One the most important factors affecting the glucose-insulin dynamics is meal intake. Hence, augmenting glucose-insulin dynamics with meal dynamics is very important. Augmented mathematical models have many patient specific physiological parameters which are difficult to estimate. Thus, there is a need to check for the identifiability of the parameters and hence there is a need to identify and estimate these parameters. The most fundamental research, which goes into automation of insulin infusion into the T1DM patient is the development of control algorithms. In this paper, a brief review on some of the important augmented models, identification, parameter estimation and existing control algorithms are presented. For the development of Artificial Pancreas System (APS) integration of glucose sensor is an important issue which in turn introduces sensor noise in the measurement, thereby leading to model imperfection. This paper further discusses the design of control algorithms in the presence of such noises and various other disturbances.