Research And Implementation Of Cardiovascular Disease Prediction System For The Elderly Based On Big Data Framework

Nowadays,the intelligent big data system has been widely used in finance,transportation,logistics and other areas.With the continuous improvement of China’s basic medical construction,the medical big data system has also been booming.Common applications of the medical big data system include personal health analysis,clinical drug research and development,chronic disease management systems and so on.Among these above,the chronic disease management systems include the recording of pathological parameters,the prediction of cardiovascular disease and other parts,so that the system can realize the storage and management of pathological parameters,and also the prediction and alarm of cardiovascular disease.With the increasing number of the elderly population in China,the aging population is accelerating.Due to the increase of age and the decline of physical function,the elderly are more vulnerable to chronic disease.However,most of the medical big data systems do not provide health services for the elderly in the community,which will lead to the ineffective management of medical data information and the low efficiency of early prevention in cardiovascular disease.In terms of this,based on the research of the existing medical big data systems in the cardiovascular disease prevention and management area,this paper aims at improving the pathological parameter data management mode,adjusting the data preprocessing and disease prediction model,also designing a cardiovascular disease prediction system for the elderly based on the big data framework.The main research contents are as follow:1.Due to the problem of low utilization rate of cardiovascular disease pathological parameter data,we establish a big data platform for these data,making use of the Spark big data framework and HIVE data warehouse components.We also achieve the purpose of sharing cardiovascular pathological information among the elderly in the community,and optimize the structure of the HIVE data warehouse by the relational algebra equivalent optimization algorithm,which can improve the efficiency of the data query.2.Due to the problem of low learning efficiency when applying BP neural network in the field of cardiovascular disease prediction,we adopt 3 ? principle and Z-score normalization algorithm to preprocess raw data.Apart from this,we are able to solve the problem of the long training time of BP neural network models in the manners of optimizing the learning rate ? and adjusting the number of neural network layers.3.We combine the big data platform we established above with the BP neural network prediction model which has been optimized as mentioned before,and then we design a cardiovascular disease prediction system for the elderly in the community based on the Spark big data framework and the improved BP neural network model.Finally,we evaluate our system,and the results show that this system can achieve the goal of effective management of the cardiovascular disease pathological parameter data.And at the same time,it can make an accurate judgment of the risk of cardiovascular disease in the elderly and assist them to have a timely understanding of their health status.