Remote Health Care System Based On Machine Learning

China’s society is beginning to enter the aging stage,the aging trend is obvious,and the population with chronic diseases is also increasing.Aging and chronic diseases in patients need to be good for a long period of time to examine and treat,but due to the limitations and other factors,such people are usually scheduled review to the hospital make a diagnosis and give treatment,the treatment effect of patients will be poor,for patients with chronic diseases,the real time monitoring of their vital signs data should be accurate otherwise it will seriously affect the doctor’s diagnosis.There are also some patients who can,t live on their own,which requires real-time monitoring in order to get better care.For the above problems,the main contents of this paper include:(1)A method of health care under non medical mode is introduced.The content of this article is to design the health monitoring system based on Android,monitor and upload human health data through physiological parameter monitoring equipment,and can monitor the physiological parameter data of monitoring object in real time,including body temperature,blood pressure,and heart rate,and alarm when the user’s health data appears abnormal.(2)In order to solve the problem of insufficient information mining and low prediction precision in multitask time series,this paper combines the supervised and semi supervised learning methods in machine learning to predict the physiological status of remote health monitoring objects.The method uses the K-means algorithm to train the same class of data clusters and uses the multitask least squares support vector machine(MTLS-SVM)to drill the historical data to predict the trend.In order to evaluate the effectiveness of the proposed method,the MTLS-SVM algorithm is compared with the K-MEANS.MTLS-SVM algorithm,and the experimental results show that the method has high predictive accuracy.(3)For long-term bedridden patients prone to phlegm problems,this paper puts forward a solution,that is,through the camera real-time video image of the patient’s throat to deal with the algorithm,proposed by the VDS algorithm to extract moving target speed characteristics and combined with DTW algorithm and SVM algorithm to classify the sequence image,To identify whether the patient has sputum obstruction symptoms,so that patients get timely and efficient care.In order to evaluate the effectiveness of this method,the experimental results show that this method has a high recognition rate compared with DTW method and SVM method.