Application of Data Mining in Medical

Medical decision support systems (MDSS) are now being used in many health care institutions across the Glove. These institutions have large amount of medical data stored in different formats and may contain relevant data that are hidden. The use of data mining is to extract hidden knowledge from a relevant data, which would be the main aim of this paper, is to show how data mining methods can be applied in medical decision support system and also to design a web based expert system that can predict heart condition using neural network. The design of the system is based on VA Medical center long beach database and collected from the University of California, Irvine (UCI) machine learning repository. After analyzing several medical decision support systems in the relevant literature, three algorithms have been identified: multilayer perceptron, decision tree and Naïve Bayes. These algorithms were tested under different configuration in order to find the best on the two medical dataset. Thereafter, a comparison was made with respect to their performance based on some set of performance metrics. The analysis is done using Waikato Environment for Knowledge Analysis (WEKA) software, on the two medical dataset which are diabetes and heart diseases database. It was found out from the analysis that was carried out; that it is quite difficult to name an algorithm tha