Abstract
This chapter aims at the presentation and comparative assessment of tools and methodologies used for the development of Personal Health Systems (PHSs) for diabetes management, early diagnosis and prevention. Medical decision support systems such as glucose prediction models, risk assessment models for long-term diabetes complications, models for early diagnosis of diabetes and closed-loop glucose controllers along with integrated systems for diabetes management are described. The outcomes of a wide range of research studies demonstrate the feasibility of providing safe, reliable and cost-effective solutions towards improving patients' quality of life through the application of PHSs. Specific limitations that prevent these systems from being fully adopted in clinical practice are highlighted, while challenges and future research directions are summarized.
TopIntroduction
Diabetes Mellitus (DM) is a group of chronic metabolic diseases characterized by elevated blood glucose levels for a prolonged period. The deregulation of glucose metabolism is due to either the insufficient insulin secretion from the pancreatic cells or impaired response of the body cells to insulin. DM is broadly classified into three main categories:
- 1.
Type 1 Diabetes Mellitus (T1DM): T1DM is an autoimmune disease caused by the destruction of insulin-producing beta cells of the pancreas resulting in the absence of insulin secretion. T1DM is usually diagnosed in children and young adults and accounts for only 5% of patients with diabetes.
- 2.
Type 2 Diabetes Mellitus (T2DM): T2DM is characterized by, either or both, insulin resistance and relative insulin deficiency. It is the most common form of diabetes accounting for at least 90% of all cases of diabetes.
- 3.
Gestational Diabetes Mellitus (GDM): GDM is characterized by high blood glucose levels during pregnancy. GDM accounts for one per 25 pregnancies worldwide. GDM usually disappears after pregnancy but it is a risk factor, for both the mother and the child, to develop T2DM in the future. Approximately half of women with a history of GDM develop T2DM within five to ten years after delivery.
DM has severe short-term and long-term complications. In particular, diabetic ketoacidosis, and hyperglycemia hyperosmolar state, are acute episodes, which may lead to diabetic coma if not treated promptly and properly. Moreover, severe hypoglycemic episodes which are caused by overdoses of administered insulin, may lead to the lost of consciousness. The excess glucose circulating through the body in the blood stream over time leads to damage of blood vessels and severe long term mortality related complications such as cardiovascular disease, diabetic neuropathy, and diabetic retinopathy.
According to the International Diabetes Federation (IDF) (2013), 382 million people (8.3% of adults) suffer from DM worldwide, while 175 million people with DM are undiagnosed. By 2035, it is estimated that 592 million people will have DM. In 2013, 5.1 million of deaths were attributed to DM while at least USD 548 billion dollars of health expenditure were caused from DM, which corresponded to a percentage of 11% of the total health spending on adults. According to the outcomes of the Diabetes Control Complications Trial, intensive glycemic control, reduces the long-term diabetes complications in T1DM (The Diabetes Control and Complications Trial Research Group, 2003). Moreover, several studies have investigated the importance of tight glycemic control for protection against the incidence of microvasclular and cardiovascular disease in T2DM (Giorgino, Leonardini, & Laviola, 2013). Intensive glycemic control involves regular glucose measurements and exogenous insulin administration, in case of insulin treated patients. To this end, latest technological advances have led to the development of Continuous Glucose Measurement Systems (CGMS) able to provide the information of glucose levels every 1 min or 5 min, and subcutaneous insulin infusion pumps (Klonoff, 2005).
Key Terms in this Chapter
Compartmental Model: Type of mathematical model with particular application in describing the transmission of materials and energies among the compartments.
Medical Decision Support Systems: Applications that provide to the users support/assistance within the context of decision making.
Patient Empowerment: Approach that puts the patient at the center of the services and provides him/her with the appropriate tools and services towards the self-management of his/her disease.
Diabetes: Chronic metabolic disease characterized by the deregulation of glucose metabolism
Data-Driven Modeling: Method used for either static or dynamic modeling, which is based on learning from the input and output data while disregarding explicit knowledge of the system’s physical behavior.
Machine Learning: Scientific discipline which is oriented to the investigation and construction of algorithms that can learn from the data.
Personal Smart Assistant: Smart mobile software agent, with the ability to perform tasks, or services, on behalf of an individual based on a combination of user input, location awareness, and the ability to communicate remotely with other devices.
Telemedicine: The combined use of telecommunication and information technologies for providing remote clinical health care.
Artificial Pancreas: Closed loop controller for the regulation of blood glucose levels in diabetic patients, through the integration of the following modules: i) insulin infusion pump, ii) algorithm for the estimation of insulin delivery rates/bolus, iii) Continous Glucose Measurement System.