Research On A Class Of Disease Analysis Model Based On Cloud Medical Treatment

With the improvement of people’s living standard, "health" and "disease" has become the focus of attention of the people, the use of computer technology for disease analysis and prediction has become one of the hot spots in the field of health care. Respiratory disease is a common disease, and it has a close relationship with meteorological and environmental factors. To study and analyze the influence of meteorological environment factors on respiratory diseases, and to establish respiratory disease predictive parsing model is of great significance. It can provide the general public with disease prevention information, provide decision basis for doctors and health institutions, to help reduce the harm caused by diseases, improve the health level of the people. this paper combined with computer technology, disease predictive parsing and medical meteorology, complete the following work.Firstly, aiming at the complexity of influencing factors and the deficiency of BP neural network in the prediction of disease, combining with the genetic algorithm and neural network integration, this paper proposes an integrated GA-BP neural network algorithm for disease prediction. This algorithm uses genetic algorithm to optimize the initial weights of the network to enhance the global search ability and increase the difference of the sub network, and the network integration theory is used to enhance the network generalization ability and the stability of the network prediction results. The experimental results show that the parallel integrated GA-BP Neural Network the disease predictive parsing model in the premise of ensuring the accuracy of prediction, with better time efficiency.Secondly, in order to meet the growth of the medical data and the r eal-time of the service, aiming at the deficiency of the integration GA-BP neural network algorithm, include time complexity and cost is high, the data growth caused by the lack of memory and so on. With the aid of Hadoop platform, this paper proposes a parallel integrated GA-BP neural network algorithm, and design a the disease predictive parsing model based on Hadoop.The experimental results show that the disease predictive parsing model based on parallel integrated GA-BP Neural Network has good network efficiency and better prediction accuracy.Thirdly,On the basis of the above,this paper further research the cloud platform architecture, combined with the characteristics of Hadoop platform, proposes a medical data processing platform based on Hadoop, the disease predictive parsing model belongs to the data analysis layer of the platform. And introduced the building process of the Hadoop platform, which provides an environment platform for the experimental verification. Then, analysis and evaluation of the disease predictive parsing model from the three aspects, which is accuracy, expansion, early warning function.