Analysis Of Sleep Stage Classification Based On Symbolic Transfer Entropy And Average Dissipated Energy

With the increasing pressure of life and work, more and more people in the world are sufferingfrom sleep problems. The quality of sleep has a great relationship with health. The result of sleepstage classification is an important indicator to measure the quality of sleep. And it is also animportant way to diagnose and treat of sleep disorders. EEG is the reaction of the brain electricalactivity. The study of sleep EEG has a great significance to improve the quality of sleep and todiagnose and treat the sleep disorders.The nonlinear analysis of sleep EEG is a hot topic. The symbolic transfer entropy betweensleep EEG and ECG and the average energy dissipation of EEG are used in sleep stageclassification. It was found that the symbolic transfer entropy and the average energy dissipationabout wake and the first stage of non-rapid eye movement sleep reflect on the changes of sleepstage. Comparatively, the symbolic transfer entropy and the average energy dissipation of wakeare larger than that of the first stage of non-rapid eye movement sleeps. The results showed thatsymbolic transfer entropy and the average energy dissipation of the two sleep stages are different.And it was confirmed by T test and multi-samples experiments.The analysis about sleep showed that brain cells and heart cells continue to be coupling as thesleep deepening. Therefore, the symbolic transfer entropy became smaller. It could clearly be seenthat the experiment results are consistent with the theory. The connection strength of neurons andthe tread about gene's imbalance and disorder continue to weaken as the sleep deepening.Therefore, the average energy dissipation decreases. It also could clearly be seen that theexperiment result is consistent with the theory. The symbolic transfer entropy and the averageenergy dissipation can apply into automatic sleep stage classification. By Multi-parameteranalysis it could achieve a higher accuracy of sleep stage classification.