本論文旨在使用血氧濃度訊號特徵及心電訊號經時頻分析所擷取之特徵，開發一睡眠呼吸中止症辨識演算法，計算睡眠呼吸中止發作次數及嚴重程度指標AHI(Apnea-hypopnea index)值。在特徵擷取上，本論文使用平滑式偽韋格納-韋立分布時頻分析取得12個心電訊號頻率隨時間變化之心律變異性特徵，以及24個由血氧飽和濃度訊號擷取出的特徵。所有的特徵再經過基於相關性之特徵選取法選出辨識效果較為良好的輸入特徵。辨識器的部分，本論文使用並比較了三種不同的辨識器，分別為倒傳遞類神經網絡、支持向量機、及貝氏分類器。最後演算法再將辨識結果轉換為發生睡眠呼吸中止事件之次數及AHI值。本研究共有69位經醫師診斷為睡眠呼吸中止症患者參與收案，蒐集每位受試者在睡眠中心入睡一晚的資料，以醫師判讀的結果作為黃金標準，並進行留一受試者交叉驗證本論文所提出之方法的有效性。結果顯示，使用Bayes classifier可得到最佳結果，其睡眠呼吸中止症次數之計算平均正確率為80.1%。其中輕中度病患之正確率為82.5%；重度病患之正確率為78.6%；極重度病患之正確率為81.6%，而演算法計算出之AHI值平均與實際值僅相差2.4。結果驗證了本演算法作為初步篩檢睡眠呼吸中止症之可行性。未來希望能將演算法結合穿戴裝置應用於居家睡眠呼吸中止症之篩檢。

This thesis proposes a sleep apnea detection algorithm based on the features extracted from the signal of blood oxygen saturation and time-frequency analysis of electrocardiography (ECG). This algorithm is used to detect sleep apnea event and calculate apnea-hypopnea index (AHI). In the feature extraction process, this study utilized a smoothed pseudo Wigner-Ville distribution (SPWVD) time-frequency analysis to generate 12 heart rate variability (HRV) features from ECG signal and 24 features from blood oxygen saturation signals. All features were then selected by a correlation-based features selection method to determine which significant features should be included to improve the classification accuracy. Using the selected features, three classifiers including backpropagation neural networks (BPNN), support vector machines (SVM), and Bayes classifiers were utilized to detect sleep apnea event and calculate AHI. A total of 69 sleep apnea patients were recruited in this study. They were arranged to stay at the sleep center for one night to collect the blood oxygen saturation signal and ECG data during sleep. The results analyzed by registered sleep technologists were served the gold standard and a leave-one-subject-out cross validation was employed to validate the proposed algorithm. The results show that Bayes classifiers reached the best performance, and the average accuracy for sleep apnea event detection was 80.1% (the accuracies of mild and moderate group, severe group and very severe group were 82.5%, 78.6% and 81.6% respectively). The average difference between the AHI obtained by the proposed algorithm and the AHI by the technologists was less than 2.5. The results have successfully validated the effectiveness of the proposed algorithm as a prescreening tool for sleep apnea. In the future, we hope the proposed algorithm can be integrated into wearable devices as an effective sleep apnea prescreening tool.

[1] F. Auger, “Time-frequency toolbox,” CNRS France-Rice University, vol. 46, 1996.

[2] A. Azarbarzin and Z. Moussavi, “Automatic and unsupervised snore sound extractionfrom respiratory sound signals,” IEEE Transactions on Biomedical Engineering, vol. 58, no. 5, pp. 1156-1162, 2011.

[3] A. Azarbarzin and Z. Moussavi, “Snoring sounds variability as a signature of obstructive sleep apnea,” Medical Engineering and Physics, vol. 35, no. 4, pp.479-485, 2013.

[4] S. Babaeizadeh, D. P. White, S. D. Pittman, S. H. Zhou, “Automatic detection and quantification of sleep apnea using heart rate variability,” Journal of Electrocardiology, vol. 43, no. 6, pp. 535-541, 2010.

[5] M. Bsoul, M. Hlaing, and T. Lakshman, “Apnea MedAssist: real-time sleep apnea monitor using single-lead ECG,” IEEE Transactions on Information Technology in Biomedicine, vol. 15, no. 3, pp. 416-427, 2011.

[6] A. Burgos, “Real-time detection of apneas on a PDA,” IEEE Transactions on Information Technology in Biomedicine, vol. 14, no. 4, pp. 995-1002, 2010.

[7] V. Carolina, “A novel algorithm for the automatic detection of sleep apnea from single-lead ECG,” IEEE Transactions on Biomedical Engineering, vol. 62, no. 9, pp. 2269-2278, 2015.

[8] L. Chen, X. Zhang, and C. Song, “An automatic screening approach for obstructive sleep apnea diagnosis based on single-lead electrocardiogram,” IEEE Transactions on Automation Science and Engineering, vol. 12, no. 1, pp. 106-115, 2015.

[9] L. Chen, X. Zhang, and H. Wang, “An obstructive sleep apnea detection approach using kernel density classification based on single-lead electrocardiogram,” Journal of Medical Systems, vol. 39, no. 5, pp. 1-11, 2015.

[10] F. Cottin, “Heart rate variability during exercise performed below and above ventilatory threshold,” Medicine and Science in Sports and Exercise, vol. 36, no. 4, pp. 594-600, 2004.

[11] S. Gami, D. O. Hodge, and R. M. Herges, “Obstructive sleep apnoea, obesity, and the risk of incident atrial fibrillation” Journal of the American College of Cardiology, vol. 49, no. 5, pp. 565-571, 2007.

[12] E. Gil, “Photoplethysmography pulse rate variability as a surrogate measurement of heart rate variability during non-stationary conditions,” Physiological Measurement, vol. 31, no. 9, pp. 1271-1290, 2010.

[13] C. Guilleminault, “Cyclical variation of the heart rate in sleep apnoea syndrome: mechanisms, and usefulness of 24 h electrocardiography as a screening technique,” The Lancet, vol. 323, no. 8369, pp. 126-131, 1984.

[14] H. H. Huang, H. L. Chan, P. L. Lin, C. P. Wu, and C. H. Huang, “Time-frequency spectral analysis of heart rate variability during induction of general anaesthesia,” British Journal of Anaesthesia, vol. 79, no. 6, pp. 754-758, 1997.

[15] S. Jasson, “Instant power spectrum analysis of heart rate variability during orthostatic tilt using a time-/frequency-domain method,” Circulation, vol. 96, no. 10, pp. 3521-3526, 1997.

[16] M. J. Lado, “Detecting sleep apnea by heart rate variability analysis: assessing the validity of databases and algorithms,” Journal of Medical Systems, vol. 35, no. 4, pp.473-481, 2011.

[17] H. Y. Liu, G. Tian, Z. J. Shi, “The comparison of time-frequency analysis methods and their applications,” Computerized Tomography Theory and Applications, vol. 24, no. 2, pp. 199-208, 2015.

[18] K. Lweesy, “New automated detection method of OSA based on artificial neural networks using P-wave shape and time changes,” Journal of Medical Systems, vol. 35, no. 4, pp. 723-734, 2011.

[19] M. O. Mendez, “Sleep apnea screening by autoregressive models from a single ECG lead,” IEEE Transactions on Biomedical Engineering, vol. 56, no. 12, pp. 2838-2850, 2009.

[20] E. Mirrakhimov, T. Sooronbaev, and E. M. Mirrakhimov, “Prevalence of obstructive sleep apnea in Asian adults: a systematic review of the literature,” BMC Pulmonary Medicine, vol. 13, no. 1, pp. 1-10, 2013.

[21] G. B. Moody, “Clinical validation of the ECG-derived respiration (EDR) technique,” Group, vol. 1, no. 3, 1986.

[22] M. Orini, “Characterization of dynamic interactions between cardiovascular signals by time-frequency coherence,” IEEE Transactions on Biomedical Engineering, vol. 59, no. 3, pp. 663-673, 2012.

[23] T. Penzel, J. W. Kantelhardt, L. Grote, J. H. Peter, A. Bunde, “Comparison of detrended fluctuation analysis and spectral analysis for heart rate variability in sleep and sleep apnea,” IEEE Transactions on Biomedical Engineering, vol. 50, no. 10, pp. 1143-1151, 2003.

[24] B. Raquel, ”Influence of running stride frequency in heart rate variability analysis during treadmill exercise testing,” IEEE Transactions on Biomedical Engineering, vol. 60, no. 7, pp. 1796-1805, 2013.

[25] K. Shafqat, “Time-frequency analysis of HRV data from locally anesthetized patients,” International Conference of the IEEE Engineering in Medicine and Biology Society, pp. 1824-1827, 2009.

[26] P. Simone, “Estimation of the power spectral density in nonstationary cardiovascular time series: assessing the role of the time-frequency representations (TFR),” IEEE Transactions on Biomedical Engineering, vol. 43, no. 1, pp. 46-59, 1997.

[27] P. Simone, “Estimation of the power spectral density in nonstationary cardiovascular time series: assessing the role of the time-frequency representations (TFR),” IEEE Transactions on Biomedical Engineering, vol. 43, no. 1, pp. 46-59, 1997.

[28] M. S. Urschitz, “Reference values for nocturnal home pulse oximetry during sleep in primary school children,” Chest Journal, vol. 123, no. 1, pp. 96-101, 2003.

[29] M. Urschitz, J. Wolff, V. Einem, P. Urshitz-Duprat, M. Schland, and C. Poets, “Reference Values for Nocturnal Home Pulse Oximetry During Sleep in Primary School Children,” Chest Journal, vol. 123, no. 1, pp. 96-101, 2003.

[30] B. Xie and H. Minn, “Real-time sleep apnea detection by classifier combination,” IEEE Transactions on Information Technology in Biomedicine, vol. 16, no. 3, pp. 469-477, 2012.

[31] 葉世彬，睡眠疾患手冊：醫生VS病人最重要的一本書，第二版，合記，台北，pp. 138-147，2013。