The Sleep Staging Method And Its Application Based On Ecg Signal

Sleep is a necessary physiological activity of human beings,and its quality is an important index to evaluate human health,and its quality is an important indicator for evaluating human health.Accurate sleep staging is a prerequisite for sleep quality assessment and the basis of sleep research.Traditional sleep staging is performed by polysomnography(PSG).The method of obtaining it is complicated,requires a lot of manpower and material resources,and needs to be performed in a laboratory or hospital,which is difficult to meet the needs of daily sleep monitoring.Therefore,finding a simple and effective sleep staging method has become an urgent problem.Based on the analysis of the relationship between heart rate variability(HRV)and sleep staging,this research proposed an automatic sleep staging method based on electrocardiogram(ECG)signals.First,the advantages of combining the Complete Ensemble Empirical Mode Decomposition with Adaptive Noise(CEEMDAN)algorithm and the Fast Independent Component Analysis algorithm(Fast ICA)algorithm to filter out noise in the ECG signal in the sleep state;Second,use the Pan-Tompkin algorithm to extract the RR interval sequence in the ECG signal and remove outliers;Then,wavelet packet decomposition(WPD)is used to decompose the RR interval sequence and extract energy features to obtain HRV features;Finally,random forest(RF)algorithm is used to model the feature data to achieve automatic sleep staging.The main research contents of this experiment are as follows:1.Combining the advantages of CEEMDAN algorithm and Fast ICA algorithm,a method for noise reduction of ECG signals based on CEEMDAN-Fast ICA is proposed,and the experimental algorithm is compared with the wavelet threshold method(WT-Subband)and S transform method.Based on the results,the advantages and disadvantages of the noise reduction algorithm are discussed.2.The time domain analysis,frequency domain analysis and non-linear analysis methods of HRV are introduced.The WPD algorithm is applied to HRV feature extraction,and an automatic sleep staging algorithm based on ECG signals is designed.3.Use the RF algorithm to train and test the pre-processed data to obtain a sleep staging model to achieve automatic sleep staging in the wake(W),non-rapid eye movement(NREM)and rapid eye movement(REM).In this experiment,CEEMDAN-Fast ICA was used to reduce the noise of the ECG signal,and the HRV feature of the RR interval sequence was extracted using the WPD algorithm,and then the sleep staging model was obtained using the RF algorithm to realize the automatic sleep staging based on the ECG signal.It can be seen from the experimental results that the accuracy rate,recall rate,F-value and accuracy rate of sleep staging of healthy subjects are0.720,0.729,0.716 and 72.9%,respectively.,F-value and accuracy are 0.780,0.777,0.843 and 77.8%,which can meet the daily sleep monitoring needs.