Medical Affiliated Hadoop Application (MAHA)

Abstract

Medical information exchange is an inevitable stage of medical informatics development. Administration features are integrated and the structure of HIE platform receives a centralized architecture which serves a enormous volume of medical information and terminates with multiple types of query and information extraction for centralized design. Cloud computing is a technological prerequisite. The purpose of this paper is that, we presents a medical information platform based on Hadoop, which is named after Medoop. The current Medoop project make essential use of HDFS to accumulate the integrated CDA records more effectively, the data content arranged in CDA record as per the cessation business queries and calculates the statistic data distributively in Map Reduce model. The scope of Medoop is to bring about a complete platform for medical data for accumulating, exchanging and using the element in Hadoop ecosystem. The project presents Medical Hadoop which consists of cloud server and estimates whether the patient is affected by diabetes or not. For prediction purpose we use Naïve Bayes algorithm. With the help of Naïve Bayes static data is provided in training phase and it trains accordingly for detection. Naïve Bayes is stipulated with normalized dataset using k-means(or any other normalization algorithm).