Studies in diabetic systems are vital for both diabetes patients and the medical industry because it is possible to use these systems to improve healthcare quality in several ways. These systems can help doctors to treat patients with all types of conditions. This research selected service oriented architecture with a decision aspect (SOAda) as the core architecture for the reason that it functions well when the domain decision is not adequately clear. This chapter demonstrates the study of the diabetes systems within service-oriented architecture and provides clinical management and advices for patients. Its aim is to (1) identify the current research state in the area, (2) assist in deriving the key problems and features with the SOAda, and (3) use these key features so that they could serve as a guide in the development of a new tool to address the decision problems in SOAda.
TopIntroduction
Method engineering techniques albeit proven effective for paradigms such as function, object and component, are pertinent to limited aspects of service orientation (Baghdadi, 2015). SOA is considered as the current and new generation of architectural style that can be utilized in computer systems. It is also considered a layered architectural style that is constructed around automated services. One of SOA’s foundations is the service inventory, and the SOA quality is affected by how the organization of the inventory is done. Within the business services layer, the services are business-oriented, while those found in other layers are seen as technically oriented. Business people play a vital role in defining business-oriented services. This includes defining business process models, corresponding services, and decision models.
Furthermore, there are service roles within SOA as well. As such, it helps to consider a service role as representative of a separation of concern. Majority of common service roles are considered to be task services and data services or information. The recognition of decision services is less common, as it represents the important business logic separation within SOA. The decision model and business decisions offer a unique way of respectfully separating business logic, possibly at a granular level. One can then roll them up into suitable decision services given all of the features of a good-quality service.
The intelligent systems that are utilized in diabetic are vital in the medical industry because they enable nurses and doctors to collect information quickly and process it in different ways to help in making diagnosis and treatment decisions.
Diabetes can be classified into (Mishra, Singh, Rana, & Shailendra, 2014):
- 1.
Type 1 Diabetes: Also known as insulin-dependent diabetes mellitus, which describes an autoimmune disease that stops the production of insulin, therefore requiring treatment by insulin injections, as well as the development of a meal plan and regular exercise.
- 2.
Type 2 Diabetes: Also known as non-insulin-dependent diabetes mellitus, which happens when tissues stop responding to insulin, and has associations with heredity and obesity and could be controllable by diet; it also makes up 90 percent of all cases, most of which are not diagnosed for years.
- 3.
Gestational Diabetes: Happens in pregnancy, usually during the 24th week. A gestational diabetes diagnosis does not indicate that diabetes was there before conception, or that the diabetes will stay after giving birth. However, following the doctor's advice about blood glucose (blood sugar) levels is important while pregnancy is being planned to ensure the health of both the mother and the baby.
Problem
Service Oriented Architecture with a Decision Aspect (SOAda) (Boumahdi & Chalal, 2014) could help in the diagnosis of diabetes patients by modelling and classifying their diabetes and providing diabetic patients with advice on how to manage their condition. This research involves the development of an “SOA” architecture design that can be used to classify diabetes with improved accuracy. Several techniques have been used in the classification of diabetes risk, and they will be described in this chapter. However, these techniques do not have the ability to clearly classify diabetes. The majority of the techniques that are given in literature offer binary classification of whether a patient is diabetic or non-diabetic. In this chapter, the computational challenge is classifying the diabetes with more accuracy, such as demonstrating the decision level from the SOAda (Boumahdi & Chalal, 2014). Instead of just classifying whether a person is diabetic or not, showing their diabetes risk levels will have more use to people as it could help diabetes patients manage their health conditions by living a proper life style based on their risks. In this chapter, the SOA architecture design reported is focused on studying diabetes classification for people that have diabetes. However, one can also apply the model to identify the risk levels for non-diabetics.