Answers are grounded in this paper and cite their sources.
Abstract—This work develops a number of automatic slow wave sleep (SWS) detection methods that employ only one or two channels of EOG/EEG signals. In addition to the reduction of signal channels, a distinct feature of the proposed approach is the introduction of a new feature set that can make the proposed approach insensitive to interpersonal differences of the physiological signals. The tested subjects include 265 and 947 persons underwent full overnight polysomnography from two different sleep centers. With 265 subjects from one center as the training set and 147 subjects from the other center as the validation set, the first part of our experiments yields SWS detection results of Kappa coefficients 0.72-0.78, sensitivity 0.77-0.90 and positive predictive value 0.73-0.82. Using the 947 subject dataset, the second part of the experiments compares the relative merits of the tested methods and investigates the impacts of SWS ratio and severity of sleep apnea on the performances of the proposed methods. Finally, our results suggest that the quality of the training set is of great importance for the development of accurate SWS detection methods.
No takes yet. Share an insight, caveat, or question.
Hang et al. (2013) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: