Development Of Sleep Quality Evaluation System Based On Multi-parameter Physiological Signals

The quality of sleep is closely related to people's academic life and physical health.Most people suffer from sleep disorders but are hard to detect.A key issue in the field of sleep medicine is how to effectively comprehensively evaluate the quality of human sleep at night.At present,the recognized standard for the evaluation of sleep quality throughout the night is still a polysomnographic monitoring system.The main problem that can be used with related commercial equipment is that there are too many sensors,which affects the overall evaluation result,and the monitoring process is cumbersome,which is likely to cause certain psychological pressure on patients.On the other hand,the automatic analysis and diagnosis functions and ease of use of such devices have major shortcomings.At the same time,the purchase and use costs are high,which is not conducive to promotion.Based on the above problems,this subject designs and implements a sleep quality evaluation system based on multi-parameter physiological signals.Based on the existing technology platform in the laboratory,it optimizes and integrates the system,and integrates multiple types of sleep-related events,the final objective and specific evaluation of human sleep quality at night.The main contents of this article include:Based on the physiological signals needed to evaluate the quality of human sleep at night,the subject proposed the development of a hardware monitoring system capable of collecting11 physiological signals simultaneously.The collected signals include: EEG,ECG,body posture,blood oxygen,and mandibular electrical,Horizontal eye movement,snoring,pulse rate,chest and abdomen breathing,nasal airflow,etc.The main tasks include hardware design,software driver and algorithm research.The hardware part includes the main control platform control connection breathing-blood oxygen integrated acquisition module and multi-channel electrophysiology integrated acquisition module;the software design mainly includes the underlying driver software design and the host computer interface display design.The main part of the underlying driver is to drive the(Microcontroller Unit)MCU's peripherals,data acquisition,preprocessing,design of some filtering algorithms,data transmission,etc;theinterface design on the(Personal Computer)PC side includes two types: display interface design and analysis interface design.The functions included in the display interface are real-time display of multi-channel physiological signals,data storage,and real-time interaction with hardware platforms.The main functions of the analysis interface include playback of the data frames of each channel,setting of parameter thresholds,output of a sleep quality evaluation report based on the doctor's recommendations.The algorithm part mainly integrates four parts of comprehensive analysis such as sleep disordered breathing events,sleep cardiovascular events,sleep EEG wakefulness events,sleep staging research,and after analyzing all kinds of events,the sleep quality is specifically evaluated.In the experimental verification,all kinds of event parameters determined by software automatic analysis are compared with those determined manually.For the cardiovascular events during sleep,the average error calculated by statistical analysis method was 0.25,and the root mean square error was 0.31.For the error analysis of the results of each index during sleep,the average of Apnea Hypopnea Index(AHI),Sleep efficiency(S),Arousal response Index(Ar I),Apnea Index(AI)and Hypopnea Index(HI)errors are 2.43,0.035,3.72,2.54 and 1.92 in turn,and the root mean square errors are 3.05,0.044,4.66,3.18 and 2.41 respectively.The accuracy of the algorithm in this article is about 94% for respiratory disturbance events,about 95% for cardiovascular events,about 85% for wakefulness events,and 86% for the classification of sleep stages.The accuracy of the comprehensive sleep quality assessment was about90%.Compared with the existing polysomnography,the accuracy of the automatic analysis of the system software and the ease of use of hardware equipment have been refined and improved to a certain extent,and have a good prospect of productization.The project also preliminary established the Shen Zhen University-Sleep Breathing Database,which includes all-night sleep data and calculation results of related parameters for 48 volunteers.In summary,the successful development of the evaluation system of this subject can assist sleep doctors in the effective screening and diagnosis of the overnight sleep state of patients with sleep disorders,and output the corresponding results report to provide a reliable reference for the evaluation of subsequent treatment effects.The preliminary establishment of the Shen Zhen University-Sleep and Respiratory Database will provide more research samples for other domestic researchers of sleep diseases,which will have a certain role in promotingthe diagnosis and treatment of sleep-related respiratory diseases in China.