Reliable sleep staging of unseen subjects with fusion of multiple EEG features and RUSBoost. (September 2021)

• Record Type: Journal Article
• Authors: Jain, Ritika; Ganesan, Ramakrishnan Angarai
• Is Part Of: Biomedical signal processing and control. Volume 70(2021)
• Publication Date: 2021-09
• Subjects: AR model -- Band power ratios -- Disorders of consciousness -- DWT -- EEG -- Higuchi fractal dimension -- Hurst exponent -- LZC -- Sample entropy -- Sleep staging; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2021.103061
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: Single/multi-channel electroencephalography-based sleep staging method is proposed. Multimodal features are employed to capture information about different sleep stages. An ensemble classifier using random undersampling with boosting technique is employed. Promising results across three different datasets and on unseen test subjects. Can be used for automatic sleep scoring to assist in the diagnosis of sleep disorders. Abstract: Extensive experiments have been carried out in this study to classify sleep EEG from three different standard databases – Sleep EDF, DREAMS and Expanded sleep EDF databases. Both two-class (sleep-awake) and multiclass classifications have been performed using a fusion of various EEG features and an ensemble classifier called random undersampling with boosting technique (RUSBoost). The results achieved using a single channel EEG are comparable or better than the state-of-the-art methods in the literature for both types of classification, on all the databases. Two-class classification is useful to determine the preferred timings for sensory stimulation of patients with disorders of consciousness. 10-fold cross-validation accuracies of 92.6% and 97.9% have been obtained on Sleep EDF database for 6-class and 2-class problems, respectively. Using Expanded Sleep-EDF dataset, the accuracies improved to 96.3% for 6-state and 99.8% for 2-state classification. For DREAMS dataset, we achieved an accuracy of 96.6% for 2-state classification. Unlike most research in the literature where performance on unseen subjects is not considered, we report classification results on the data from unseen test subjects using both 50%-holdout and leave-one-out cross-validation approaches. Similar results were achieved using both validation techniques for different datasets emphasizing the reliability of our method. These results are very crucial for the method to be applicable for clinical use.