In the health care industry, there are lot of examples that are supported by the decision support system with the growth of health care industry in the medical field. In 1994, Bangemann EU report explained the example of implementing the decision support system which foresees that it is used in the cost-saving of health care issues in best services provided to the patients. The report shows how the industry improves the services in better way, which puts the patients at maximum ease to increase the efficiency of the dataset system. In Finland, the ministry of health enforces a strategic plan according to the decision support system, which was beneficial for the cost-effectiveness and economical in the industry of health to care the social services in the visions of implementation of the cost of treating the issues. There are several other examples in the health industry which used decision support system in problem-solving. There is example of Blood glucose screening level is the tool of major support systems used in the health care organization, major obesity, and bad eating habits are the main reason for having severe diseases like Hypertension, diabetes, and cardiovascular diseases. The screening rates of the diabetic and known diabetic patient in 3 years is 8388.for lipid screening between the overweight persons and normal-weight persons overall, 70% of persons were screened. After having a lipid screening test of the diabetic patient, it is concluded that 85% of persons over the age of 45 had a chance of increasing blood glucose levels. It is the major decision support system in the health organization (R. Bose, 2003).

2.    Discuss the major algorithms used for decision support systems for the healthcare industry

The algorithms which are used for the decision support systems for the healthcare industry are discussed in this report. The scientists propose new technique, the selection of the feature that is named locally temporal feature selection (LTFS), which is a local selection technique for the one query patient, which recursively determines the most relevant feature based on the recent values as well as the features subset. On the other side, the next step test is recognized based on that information that is noted or appeared such as the previous results as well as symptoms. When the treatment starting point is reached such as the threshold probability, which identifies that the patient has the disease, the algorithm will stop. To select a test, the criteria are used for this purpose including the relevance of the associated cost as well as the feature. Furthermore, I implement the criteria in the cost-based objective functions as well as speed based functions (K. B. Matthews, et al., 1999).

Three datasets are runs against ID3 algorithms, and results are shown in the figures. Furthermore there is also CART algorithm, which is based on the classification of the trees used in the test and produced non-parametric techniques, which gives results in regression trees and describes the nature of the variable as it depends variable or independent. Trees are formed according to the data given to the datasets in the equation and outcomes are measured wither particular modeling. Rules of testing also based on the nature of how it split under the various observations. Once the rule of splitting the variables is made it couldn’t be changed with the time as it based on the rules of the dataset. When CART is detected to gain more to meet the rules, which are growing in the possible outcomes, it could predict better results about the disease. CART algorithm includes:

·         To solve a big tree problem in testing.

·         To incorporate automatic data validation for test.

·         Provide a new method to deal with the missing values.

The outcomes of the CART algorithms for the health care department could be explained in the values which could be high or less in the way to measure the disease. The algorithms are used to classify to get a better understanding of the method to use in specific dataset. CART algorithms are used in the heart disease dataset which is more complex in satisfactory level. CART shows almost 80% accuracy in the dataset which is examined through CART. All the algorithms are designed to get conquer approach in building a perfect decision support system. There are also C4.5 algorithms which are used in medical health care. Another algorithm was used in the system with the name of the Lazy Learning Algorithm, which is also known as memory-based learning, instance-based learning or case-based learning that makes a prediction model, particularly for the query case. Algorithms could be used in the prediction of the test of different diagnose problem in the health industry. More algorithms used in the health sector are ID3 Algorithms, which was first introduced by the J.R Quinlan in 1970s. These algorithms choose different attributes to gain data about a particular disease (E. Aktaş, et al., 2007).  

References of Industry overview / Scenario of The decision support system in healthcare

E. Aktaş, Ülengin, F. & Şahin, Ş. Ö., 2007. A decision support system to improve the efficiency of resource allocation in healthcare management.. Socio-Economic Planning Sciences, 41(2), pp. 130-146..

K. B. Matthews, Sibbald, A. R. & Craw, S., 1999. Implementation of a spatial decision support system for rural land use planning: integrating geographic information system and environmental models with search and optimisation algorithms.. Computers and electronics in agriculture, 23(1), pp. 9-26..

R. Bose, 2003. Knowledge management-enabled health care management systems: capabilities, infrastructure, and decision-support.. Expert systems with Applications, 24(1), pp. 59-71..