Research And Implementation Of DS Evidence Theory In Mobile Medical Platform

With the improvement of living standard,people expect to get better medical service.The rapid development of computer technology makes it has been widely used in all aspects of the medical field,however,the computer technology is mainly used for patient medical information collection,display,storage,etc.,the diagnosis of the disease are still largely by doctors to identify.The use of computer to assist the diagnosis and prediction of diseases and realize intelligent medical treatment is at the beginning stage of development.In addition,the rapid change of the Internet makes mobile medical research and application development platform,data processing,mobile technology,mobile medical platform of computer network in one can to a certain extent,solve the huge medical patients,doctor-patient communication at any time,historical data could not effective accurate storage and so on the problems existing in the traditional medical mode.Mobile medical platform is a complete set of data processing system including mobile phone mobile client,background management platform and database.Patients can communicate with doctors and nurses through mobile client operation to check their medical records.The nurse uploaded the patient's medical records and other data through the background management platform;The doctor coming through the data on the mobile client can effectively in real time to determine the user's health,automatically by the client's disease prediction algorithm for judging the patient's condition and prediction.Therefore,the application of computer technology in the diagnosis and prediction of diseases is of great significance for the early detection of diseases,early treatment and reduction of misdiagnosis.This paper first introduces the basic concepts and theories of BP neural network and DS evidence theory,and summarizes their respective characteristics and the theory of algorithm implementation.Then the paper introduces the design and implementation of the remote rehabilitation system for children with chronic nephropathy,and introduces each function module in detail.And then,In view of the characteristics of BP neural network and DS evidence theory,this paper proposes a model of combining BP neural network with DS evidence theory for disease prediction.Using BP neural network model,the model get diseases forecasting model,using the real data to forecast the disease according to the characteristics of the model,the results with the DS evidence theory to carry on the synthesis,synthesis of the results as the final prediction results.The algorithm integrated the DS evidence theory to the advantage of the subjective factors as a source of evidence,will more than the judgment results of illness doctors to predict fusion results,to improve the accuracy of prediction result,the algorithm makes up for the further the shortcoming of BP neural network prediction accuracy is not high.In chronic kidney disease(CKD)as a case study,to solve the above innovation of this article children with chronic kidney disease in remote rehabilitation system,according to the result of prediction for the doctor to suggest rehabilitation guidance.The experimental results show that the introduction of DS evidence theory can comprehensively improve the accuracy of the prediction of chronic nephropathy,and the robustness of the algorithm is better.