Parallel Bayesian Network Modelling for Pervasive Health Monitoring System

Nowadays patients are experiencing pervasive health monitoring service. In order to improve the quality of service, managers and clinicians need to analyze plentiful data collected by medical information system. In this way, because they can acquire useful knowledge, wise decision will be made and the treatment and prevention will become more effective. Since Heart Disease (HD) is considered as one of the main reasons of death for adults, heart disease analysis deserves more attention. This paper uses Bayesian networks to analyze heart disease and presents a method to conform the sequence of network nodes from sample dataset. This method overcomes the limitation of traditional algorithms, which require experts of medical field give the order of network nodes. In addition, in order to shorten the analysis time, a parallel optimization technique is adopted to accelerate the establishment of HD diagnosis model over large amounts of data. Experiments show that the proposed method can improve the accuracy of the modeling and shorten the modeling time to some extent.