Predictive pattern analysis using SOM in medical data sets for medical treatment service

This paper proposes a new method of patterns analysis using SOM in medical data sets for medical treatment service under ubiquitous computing environment which is required by real time accessibility and agility. In this paper, it is necessary for us to classify disease patterns in the medical historical record to join the information of patient, using SOM neural network with input vectors of different features, disease code, input factors in order to take the medical treatment service in medical data sets, to reduce patients´ search effort to get the information of diagnosis for recovering their health and to improve the rate of accuracy. To verify improved performance, we make experiments with dataset collected in medical center.