Design And Implementation Of Smart Homecare Bed Based On Health Data Mining

With the deepening of the aging of China's population,the proportion of the elderly population to the total population of the country continues to grow.The issue of life care and health care for the elderly is receiving increasing attention.The issue of old-age care has also become a hot issue in society.In recent years,under the background of China's policy to encourage and support the old-age industry,it has become a trend to realize the intelligent and informatization of old-age services.With the development of the "Big Data+" model and related technologies,the concept of smart pension has gradually emerged and popularized.The system is based on the electric old-age bed system,which can effectively reduce the daily nursing difficulty of the nursing staff,and assist the elderly to achieve the actions of starting the back,bending the legs,sideways,and leaving the bed.Although such electric nursing beds have been well applied,due to the lack of integration with new sensing devices,such as vital signs sensors and smart wearable devices,continuous monitoring of physiological parameters of the elderly is not possible,and the current body of the subject cannot be provided.The assessment of health status cannot predict the physical health of the elderly.In order to solve the problems of simple function and backward technology of domestic electric homecare bed,a smart homecare bed based on health data mining is proposed.Firstly,the human heartbeat signal and the respiratory signal are collected by the non-contact vital sign sensor,and the weight signal of the human body is collected by the pressure sensor,and the collected physiological parameters of the human body are transmitted to the remote server based on the wireless module.Then,by comparing the efficiency of classification and clustering methods in heart rate and respiratory rate data mining,the health data mining algorithm is implemented based on k-Means clustering.Once more,considering the limitation that the final clustering result of the traditional K-means clustering algorithm is greatly affected by the initial value selection,a maximal minimization initialization method is introduced,which is compared with the improved algorithm by the results of the typical algorithm and the improved algorithm.The clustering center error is smaller,the average error is decreased from 1.4769% to 0.0767%,and the clustering result is closer to the standard value.Finally,combined with the characteristics of the time series data set,the gray prediction method is introduced,and the gray GM(1,1)prediction method is used to predict the human physiological data and provide data reference for the user's physiological health prediction.The experiment proves that the smart homecare bed system based on health data mining proposed in this paper can not only solve the problem of simple function and backward technology of traditional electric homecare bed,but also provide accurate health data guarantee for subsequent health care.