Sleep stage classification using single-channel EOG

• Published in: Computers in biology and medicine Vol. 102; pp. 211 - 220
• Main Authors: Rahman, Md Mosheyur, Bhuiyan, Mohammed Imamul Hassan, Hassan, Ahnaf Rashik
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.11.2018; Elsevier Limited
• Subjects: Accuracy; Algorithms; AR model; Autoregressive models; Autoregressive processes; Classification; Datasets; Discrete Wavelet Transform; Discrete Wavelet Transform (DWT); EEG; Electrodes; Electroencephalography; Electrooculogram (EOG); Entropy; Experts; Internal Medicine; Methods; Neighborhood Component Analysis (NCA); Other; Physiology; Random forest; Random Under Sampling Boosting (RUSboost); Sleep; Sleep disorders; State of the art; Support Vector Machine; Support vector machines; Variance analysis; Wavelet analysis; Wavelet transforms; AR model; Electrooculogram (EOG); Discrete Wavelet Transform (DWT); Neighborhood Component Analysis (NCA); Random forest; Support Vector Machine; Random Under Sampling Boosting (RUSboost)
• DOI: 10.1016/j.compbiomed.2018.08.022
• ISSN: 0010-4825; 1879-0534

Sleep stage classification is an important task for the timely diagnosis of sleep disorders and sleep-related studies. In this paper, automatic scoring of sleep stages using Electrooculogram (EOG) is presented. Single channel EOG signals are analyzed in Discrete Wavelet Transform (DWT) domain employing various statistical features such as Spectral Entropy, Moment-based Measures, Refined Composite Multiscale Dispersion Entropy (RCMDE) and Autoregressive (AR) Model Coefficients. The discriminating ability of the features is studied using the One Way Analysis of Variance (ANOVA) and box plots. A feature reduction algorithm based on Neighborhood Component Analysis is used to reduce the model complexity and select the features with highest discriminating abilities. Random Under-Sampling Boosting (RUSBoost), Random Forest (RF) and Support Vector Machine (SVM) are employed to classify various sleep stages for 2–6 stage classification problem. Performance of the proposed method is studied using three publicly available databases, the Sleep-EDF, Sleep-EDFX and ISRUC-Sleep databases consisting of 8, 20 and 10 subjects respectively. The proposed method outperforms the state-of-the-art EOG based techniques in accuracy. In addition, its performance is shown to be on par or better than those of various single channel EEG based methods. An important limitation of existing sleep detection methods is the low accuracy of the S1 sleep stage classification for which the proposed method using the RUSBoost classifier gives a superior accuracy as compared to those of EOG and EEG based techniques. •A single-channel Electrooculography based automated sleep scoring method is proposed.•Discrete Wavelet Transform is employed for time-frequency decomposition of the Electrooculography signals.•A feature reduction technique based on Neighborhood Component Analysis is used to reduce the number of features.•The performance of the proposed scheme is promising.•The proposed scheme can be used for sleep stage detection in ambulatory sleep studies and mobile applications.