Research And Implementation Of Family-oriented Intelligent Medical Platform

With the rapid economic and social development and the widespread popularization of health care awareness in China,people's awareness and demand for health care are also deepening and improving.Family is the most basic unit for people to get a foothold in society.Intelligent medical care aimed at the problem of insufficient medical resources and serious aging population,integrating health care services and information and communication technologies,extracting value information from a large amount of medical information data based on big data and artificial intelligence technology,to implement a series of monitoring measures based on data.It extended the health care scenario to families,and promoted the emergence of family-based health care model.This thesis designs and implements a family-oriented intelligent medical platform,which lays a foundation for users to enjoy health care services at home,and studies the prediction algorithm of hypertension based on convolutional neural network,which provides a new research idea for the subsequent medical data processing algorithms.Aiming at the difficult problem of medical data collection,display and storage service nedded in the intelligent medical platform,an intelligent medical platform is designed,which consists of data collection system,client and server.The data collection system implements the functions of data collection and forwarding,and is divided into a data collection module and a data forwarding module.The user monitors the surrounding environment information and the physiological medical information by using the sensor and the smart medical device at home.The data collecting module receives the data through ZigBee protocol or BLE protocol,processes it and sends it to the data forwarding module.Then the data is forwarded to the client based on the Wi-Fi protocol.Based on the Android technology development,the client implements the basic functions including registering the login account,receiving and displaying the data forwarded by the data collection system,and uploading the data to the server.In addition,considering the needs of family users to understand each other's health,it also designed to add friends and relatives and data query function.In addition to viewing historical data,users can also view relatives and friends' user data,which facilitates mutual care among family members;and designed the disease prediction function,which can predict the user's physical condition based on the collected medical information.The server is built on the Django framework and the MySQL database to implement the function of data storage and query feedback.Finally,this thesis studies the hypertension prediction algorithm based on convolutional neural network.The dataset is derived from the MIMIC database.The structure of the convolutional neural network has been adjusted for the characteristics of the eight physiological characteristics data used in this thesis,which takes into account the richness of data characteristics and the trend of data changes over time.Eventually,the model CNN-HP realizes on the prediction effect improved,and the classification accuracy rate is 90.15%,which is better than the six classical machine learning methods.