Affiliations: [a] School of Engineering, Sun Yat-Sen University, Guangdong, China | [b] Jinan University, Guangdong, China | [c] Sleep-Disordered Breathing Center of the 6th Affiliated Hospital of Sun Yat-Sen University, Guangdong, China | [d] Department of Mechanical, Electro-Mechanical Engineering National Sun Yat-Sen University, Kaohsiung, Taiwan
Correspondence:
[*]
Corresponding author: Yuxi Luo, School of Engineering, Sun Yat-Sen University, Guangdong 510006, China. Tel.: +86 13631462576; E-mail: [email protected].
Abstract: Background:Healthy sleep can be characterized by several stages: wake, light, SWS, and REM sleep. The clinical experts find that the breath of subjects is different in these sleep stages, but such observation is lacking data supporting, The statistical research about investigating breathing patterns during sleep process will be helpful for the sleep and breathing domain. Objective:The objective of the paper is to statistically analyze the respiratory characteristics during different sleep stages. Methods:Firstly, we calculated the mean value and standard deviation of respiratory rates of these stages, in which the respiratory rates were obtained by the autocorrelation method. Then the detrended fluctuation analysis (DFA) algorithm was applied to analyze long-range correlation of respiratory rates of sleep stages. Results:The mean and standard deviation of respiratory rates are wake: 16.62 ± 2.43 cycles per minute (CPM), light: 15.15 ± 1.53 CPM, SWS: 15.06 ± 0.96 CPM and REM: 16.37 ± 2.03 CPM, respectively. The scaling exponent applied by detrended fluctuation analysis (DFA) algorithm reached about 0.7 for each stage. Conclusion:Results of the mean and standard deviation of respiratory rates show that different sleep stages lead to different autonomic regulations of breathing and exhibit different respiratory rates and fluctuations. And the DFA results demonstrate that respiratory rates are all long-range correlated in these stages although they lead to different fluctuation.
