睡眠呼吸中止症是一種因呼吸道反覆性的塌陷造成呼吸道阻塞，進而導致停止呼吸的疾病。關於睡眠呼吸中止症的診斷，醫生會請病人至醫院進行睡眠檢查，透過多重睡眠圖儀器以及病人的打呼與咳嗽進行分析。因睡眠呼吸中止症之病人與日俱增以及醫院睡眠中心病床不足，造成大量排隊的狀況，導致嚴重的病患無法即時檢查與治療。因此我們提出一個打呼與咳嗽偵測機制，讓病患可使用該機制進行長期之夜間睡眠打呼與咳嗽偵測，並且將打呼與咳嗽之數據量化，協助醫生進行病症診斷，由醫生判斷該病患是否需要進行睡眠檢查，可以疏通大量排隊之人群。

此偵測機制包括三個部分，第一部分，將病人整夜睡眠聲音資料進行獨立事件切割；第二部份，我們將時域上的訊號利用複利葉轉換轉至頻域訊號，分別進行打呼與咳嗽聲音之特徵擷取；第三部份，利用支持向量機(Support Vector Machine)與隱藏馬可夫模型(Hidden Markov Model)為建立偵測機制之模型，進行夜間打呼與咳嗽聲音偵測。

Obstructive Sleep Apnea (OSA) is a respiratory tract obstruction caused by recurrent respiratory tract collapse, then leading to stop breathing disease. About OSA diagnosis, the doctor will ask the patient to the hospital for a sleep examination, using Polysomnpgraphy and patient’s snore and cough for analysis. Because the number of OSA patients has been increased and hospital beds are not enough, resulting in a large number of queued condition that causes severe patient cannot be immediate examination and treatment. Therefore, we propose a mechanism to detect snore and cough, patients can use this mechanism for snore and cough detection in long-term sleep at night. The snore and cough quantitative data help doctors to diagnose diseases, and doctors determine whether the patient need for sleep examination or not. The mechanism can eliminate a lot of patients queuing.

The detection mechanism include three parts. First, the patient all night sound data segments to independent events. Second, we change time domain signal to frequency domain signal by Fourier Transform, and extract features from snore and cough respectively. The last one, we use Support Vector Machine and Hidden Markov Model for establishing the detection mechanism for snoring and coughing sound detection at night.

[1]“台灣飛利浦、台灣睡眠醫學學會與資深藝人趙舜聯手呼籲治療的重要性”, from http://www.newscenter.philips.com/tw_zh/standard/about/news/2013/20130319.wpd#.Ue9I740yQs4

[2]嘉義長庚睡眠醫學中心,”睡眠相關問題”, from
http://www1.cgmh.org.tw/sleepcenterjia/index.files/Page697.htm

[3]王棨德醫師,”鼾聲如雷: 打呼與睡眠呼吸中止”, from
http://ctwang.pixnet.net/blog/post/22022995-%E9%BC%BE%E8%81%B2%E5%A6%82%E9%9B%B7%3A-%E6%89%93%E5%91%BC%E8%88%87%E7%9D%A1%E7%9C%A0%E5%91%BC%E5%90%B8%E4%B8%AD%E6%AD%A2

[4]M. Cavusoglu , M. Kamasak , O. Erogul , T. Ciloglu , Y. Serinagaoglu
and T. Akcam, “An efficient method for snore/nonsnore classification
of sleep sounds, ” Physiol. Meas., vol. 28, pp. 841–853, 2007.

[5]Ali Azarbarzin , and Zahra M. K. Moussavi, “ Automatic and Unsupervised Snore Sound Extraction From Respiratory Sound Signals, ” IEEE transactions on Biomedical Engineering, vol. 58, NO. 5, 2011.

[6]W. D. Duckitt, S. K. Tuomi, and T. R. Niesler, “Automatic detection,
segmentation and assessment of snoring from ambient acoustic data,”
Physiol. Meas., vol. 27, pp.1047–1056, 2006.

[7]A. S. Karunajeewa, U. R. Abeyratne, and C. Hukins, “Silence—
breathing—snore classification from snore-related sounds,” Physiol.
Meas., vol. 29, pp. 227–243, 2008.

[8]SH. Shin, T. Hashimoto, S. Hatano, “Automatic Detection System for Cough Sounds as a Symptom of Abnormal Health Condition, ” IEEE Transactions on Information Technology in Biomedicine, Vol.13(4), pp. 486 – 493, 2009

[9]S. Matos, S.S. Birring, I.D. Pavord, and D.H. Evans, “Detection of Cough Signals in Continuous Audio Recordings Using HMM,” IEEE Trans. Biomedical Eng., vol. 53, no. 6, pp. 1078-1083, 2006.
[10]Roger Jang (張智星),"Audio Signal Processing and Recognition", from
http://mirlab.org/jang/books/audioSignalProcessing/

[11]Control Chart(管制圖),from http://www.SPConline.net

[12] Pearson, K., "On Lines and Planes of Closest Fit to Systems of Points in Space", Philosophical Magazine 2 (11): 559–572.

[13]V. Vapnik, “Statistical Learning Theory, “Wiley, New York, 1998.

[14]Roger Jang (張智星),"Data clustering and Pattern Recognition," from
http://mirlab.org/jang/books/dcpr/

[15]L. E. Baum, T. Petrie, G. Soules, and N. Weiss, "A maximization technique occurring in the statistical analysis of probabilistic functions of Markov chains,"Ann. Math. Statist., vol. 41, no. 1, pp. 164–171, 1970.

[16]Qi Wang, Lei Wei, Rodney A. Kennedy , "Iterative Viterbi Decoding, Trellis Shaping, and Multilevel Structure for High-Rate Parity-Concatenated TCM," IEEE Transactions on communications,vol.50,pp48-55

[17]Markov model, from http://www.csie.ntnu.edu.tw/~u91029/HiddenMarkovModel.html

[18]MacQueen, J. B. ,"Some Methods for classification and Analysis of Multivariate Observations,". Proceedings of 5th Berkeley Symposium on Mathematical Statistics and Probability 1. University of California Press. pp. 281–297

[19]k-means clustering, from http://en.wikipedia.org/wiki/K-means_clustering