%0 Journal Article
%A LI Jin-bao
%A Lü Xing-feng
%T A Method of Detecting Sleep Apnea Using Random Forest
%D 
%R 10.13190/j.jbupt.2019-255
%J Journal of Beijing University of Posts and Telecommunications
%P 64-70
%V 43
%N 5
%X To solve the problem that various respiratory signals in polysomnography make the detection of sleep apnea complicated and affect patients' sleep, a method of automatic sleep apnea detection using random forest is proposed. The energy and marginal spectrum distribution of sleep apnea is significantly different from that of normal sleep after Hilbert-Huang transform. By extracting the relevant frequency domain features, combining with the time domain features, the random forest method in machine learning method is used to detect sleep apnea, which effectively reduces the detection complexity and improve the accuracy. Experiments show that this method is more convenient and accurate than the existing method, more suitable for home environment,and has a wide range of application prospects.
%U https://journal.bupt.edu.cn/EN/10.13190/j.jbupt.2019-255