The Research And Development On Human Health Monitoring And Evaluation Platform

Human is currently facing high morbidity threats of chronic disease from the speeding up of life rhythm and the increasing life stress,and more and more people are suffering from chronic diseases.Therefore,it is particularly important to monitor people's health.If people's health status is changing from good to bad,then through daily monitoring,people can be alerted in advance and they will be advised to go to hospital for medical examinations to prevent the deterioration of health conditions.The traditional health monitoring methods are difficult to implement because of their complicated operations.Fortunately,the current popularity of smart phone,the rapid development of image processing technology and the development of data processing capabilities make it possible of achieving measuring human physiological parameters and monitoring health condition by an android phone.The non-contact mobile phone software for measuring human physiological parameters in this study has the characteristics of low cost and simplicity.Firstly,this paper introduces the development of photoplethysmography technology and the description of physiological measurement products based on PPG technology.After a comprehensive analysis,the feasibility and innovation in our design are expounded.At the last part of this chapter,we give the research content and structure of this article.Secondly,this article outlines the overall framework of the health monitoring platform which includes 3 parts: the mobile phone client;the server and the the health assessment model based on Hidden Markov Model(HMM).And we also list out the functions that the system needs to implement and the related technologies we needed.Then,we introduce the principle of the pulse signal acquisition using mobile phone,and we also introduce the denoising method.After denoising to signal,four kinds of human basic physiological parameters such as heart rate,respiration rate,mean pressure and blood oxygen were extracted from the signal.Next,We introduce the basic theory of Hidden Markov and the health assessment model.We also verify the accuracy and rationality of the established model.During the model building process,firstly,we acquire the pulse signal from the database and extract 38 time-frequency domains and wavelet features from the signal.The 38 features are merged with physiological features to form a high-dimensional matrix.Then the high-dimensional feature space vector is mapped to the low-dimensional space by isomap,and the sensitive features are acquired.Finally,these sensitive features are introduced into the HMM to assess human health condition.In the experimental verification section,the final results of the 8 sets data were analyzed and compared.After then,in this chapter,it mainly consists of 3 parts.The first part mainly describes the design and implementation of each module in client.The second part introduces the construction of the server-side platform.The third part we test and analyze the accuracy of the system.And we compare the results obtained by the system with results measured by the medical device.Finally,this article has built a platform that can achieve measuring human physiological parameters and monitor health condition successfully.Simultaneously,we summarize the research work and give out the future endeavor orientation in our research.