本論文主旨在於藉由血氧單一生理訊號，經由特徵擷取並執行特徵選取後，篩選出有意義之血氧特徵，搭配分類器快速篩檢受測者患有呼吸中止症之潛在可能及其嚴重程度，以提供受測者是否需要做進一步就診之參考。呼吸中止事件發生之時間長度相較於整夜睡眠時間只佔很小一部分，發生呼吸中止事件之時間相對於無呼吸中止事件之時間，屬於一種不平衡資料。一般傳統的辨識器，對於不平衡資料的類別預測，雖然可以獲得高正確率，但敏感性(Sensitivity)卻會降低，導致重要的資料無法被正確的預測出來。因此，我們利用成本敏感學習(Cost-sensitive learning)中的MetaCost算法，根據成本最優分類為準則，為訓練數據集重新標記，使得演算法能夠提高敏感性(Sensitivity)並符合應用需要；此外，為使得演算法分類正確率提升，本論文以Bagging(Classification tree)為分類器。最後，以實驗結果驗證本論文提出基於血氧飽和度及成本敏感學習法則之睡眠呼吸中止事件偵測演算法應用於快速篩檢受測者患有呼吸中止症之潛在可能及嚴重程度之高度的可行性與有效性。

This thesis presents a sleep apnea detection algorithm based on blood oxygen saturation signal and cost-sensitive learning. The proposed algorithm consists of a feature generation process from oxygen saturation signal, a feature selection process based on correlation-based feature selection, a MetaCost learning and a Bagging classification. The MetaCost learning is utilized because the sleep apnea events are imbalanced data sets in nocturnal sleep data sets. Using MetaCost learning can prevent low-sensitivity results in classification. Bagging is a machine learning ensemble algorithm designed to improve the accuracy by integrating multi-weak learning classifier. The proposed algorithm was validated through K-fold cross validations based on the data collected from the St. Vincent's university hospital / university college Dublin sleep apnea (UCD) database and a sleep center in Southern Taiwan. The receiver operating characteristic (ROC) is used to evaluate the cross-validation results of the proposed algorithm. The ROC results obtained from the cross-validation are higher than 0.86 which stands for high accuracy. In the future, we will focus on the implementation of the proposed algorithm into wearable oxygen saturation sensors.

[1] A. Meoli, K. Casey, R. Clark, J. Coleman, R. Fayle R. Troell and Conrad, “Hypopnea in Sleep-disordered Breathing in Adults,” Sleep, vol. 24, no. 4, pp. 469-470, 2001.
[2] A. Dhawan, “Medical image analysis,” Wiley-IEEE Press, vol. 31, 2011.
[3] A. Burgos, A. Go˜ni, A. Illarramendi and J. Bermudez, “Real-time Detection of Apneas on a PDA,” IEEE Trans. Inf. Technol. Biomed., vol. 14, no. 4, pp. 995-1002, 2010.
[4] B. Xie and H. Minn, “Real-Time Sleep Apnea Detection by Classifier Combination,” IEEE Trans. on Information Technology in Biomedicine, vol. 16, no. 3, pp. 469-477, 2012.
[5] C. Guilleminault, S. Connolly, R. Winkle, K. Melvin, and A. Tilkian, “Cyclical variation of the heart rate in sleep apnoea syndrome. Mechanisms, and usefulness of 24 h electrocardiography as a screening technique,” Lancet, vol. 323, pp. 126-131, 1984.
[6] D. Alvarez, R. Hornero, D. Ab´asolo, F. Campo and C. Zamarr´on, “Nonlinear Characteristics of Blood Oxygen Saturation from Nocturnal Oximetry for Obstructive Sleep Apnoea Detection,” Physiol. Meas., vol. 27, pp. 399-412, 2006.
[7] D. W. Hudgel, R. J. Martin, B. Johnson, and P. Hill, “Mechanics of the respiratory system during sleep in normal humans,” J. Appl. Physiol., vol. 56, pp. 133–137, 1984.
[8] D. Liu, X. Yang, G. Wang, J. Ma, Y. Liu, C.-K. Peng, J. Zhang, and J. Fang, “HHT based cardiopulmonary coupling analysis for sleep apnea detection,” Sleep Medicine, vol. 13, pp. 503-509, 2012.
[9] F. Lopez-Jimenez, F. Sert, A. Gami and V. Somers, “Obstructive Sleep Apnea. Implications for Cardiac and Vascular Disease,” Chest, vol. 133, pp. 793-804, 2008.
[10] I. Ayappa, B. Rapaport, R. Norman and D. Rapoport, “Immediate Consequences of Respiratory Events in Sleep Disordered Breathing,” Sleep Medicine, vol. 6, no. 2, pp. 123-130, 2005.
[11] J. Webster, “Design of Pulse Oximeters,” Medical Physics and Biomedical Engineering, 1997.
[12] J. Epstein, D. Kristo and P. Strollo, “Clinical Guideline for the Evaluation, Management and Long-term Care of Obstructive Sleep Apnea in Adults,” Journal of Clinical Sleep Medicine, vol. 5, pp. 263-76, 2009.
[13] J. Solà-Soler, R. Jané, J. Fiz, J. Morera, “Pitch Analysis in Snoring Signals from Simple Snorers and Patients with Obstructive Sleep Apnea,” proceedings of the second joint EMBS/BMES conference, vol. 2, 2002.
[14] J. Hanley and B. McNeil, “The Meaning and Use of the Area Under a Receiver Operating Characteristic (ROC) Curve," Radiology, vol. 143, pp. 29-36, 1982.
[15] J. Wong, W. Lu, K. Wu, M. Liu, G. Chen and C. Kuo, “A Comparative Study of Pulse Rate Variability and Heart Rate Variability in Healthy Subjects," Journal of clinical monitoring and computing, vol. 26, pp. 107-114, 2012.
[16] K. Ramakrishnan, D. Scheid, “Treatment Options for Insomnia,” American Academy of Family Physicians, vol. 76, pp. 517-526, 2007.
[17] L. Breiman, “Bagging Predictors,” Mach. Learn., vol. 24, no. 2, pp. 123-140, 1996.
[18] M. Malik, A. J. Camm, R. E. Kleiger, A. Malliani, A. J. Moss, and P. J. Schwartz, “Heart Rate Variability : Standards of Measurement, Physiological Interpretation, and Clinical Use,” European Heart Journal, vol. 17, pp. 354-381, 1996.
[19] M. Hall, “Correlation-based Feature Subset Selection for Machine Learning,” Ph.D. dissertation, Univ. Waikato, Hamilton, New Zealand, 1998.
[20] M. O. Mendez, A. M. Bianchi, M. Matteucci, S. Cerutti, and T. Penzel, “Sleep apnea screening by autoregressive models from a single ECG lead,” IEEE Trans. on Biomedical Engineering, vol. 56, no. 12, pp. 2838-2850, 2009.
[21] M. Pohl, J. Leehan, “Frequency Analysis of Simulated Snoring Sounds Using Burg’s Estimator,” IEEE-EMBC and CMBEC Theme 4: Signal Processing, vol. 2, pp. 949-950, 1995.
[22] 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, vol. 123, no. 1, pp. 96-101, 2003.
[23] M. Adnane, Z. Jiang and Z. Yan, “Sleep–wake Stages Classification and Sleep Efficiency Estimation Using Single-lead Electrocardiogram,” Expert Systems with Application, vol. 39, pp. 1401-1413, 2012.
[24] O. Richard, P. Hart and D. Stork, “Pattern Classification,” Wiley-Interscience, 2001.
[25] P. Levy, J. Pepin, C. Blanc, B. Paramelle and C. Brambilla, “Accuracy of Oximetry for Detection of Respiratory Disturbances in Sleep Apnea Syndrome,” Chest, vol.109, pp. 395-399, 1996.
[26] P. Chazal, C. Heneghan and W. McNicholas, “Multimodal Detection of Sleep Apnea Using Electrocardiogram and Oximtery Signals,” Philosophical Transactions of The Royal Society, vol. 367, pp. 369-389, 2009.
[27] P. Domingos, “MetaCost: A General Method for Making Classifiers Cost-sensitive,” Proceedings of the fifth ACM SIGKDD international conference on Knowledge discovery and data mining, pp. 155-164, 1999.
[28] P. Chan and S. Stolfo, “Toward Scalable Learning with Non-uniform Class and Cost Distributions,” in Proceedings of the 4th International Conference on knowledge Discovery and Data Mining, vol. 164, pp. 168, 1998.
[29] R. Hornero, D. Alvarez, D. Abasolo, F. Campo and C. Zamarron, “Utility of Approximate Entropy From Overnight Pulse Oximetry Data in the Diagnosis of the Obstructive Sleep Apnea Syndrome,” IEEE Trans. on Biomedical Engineering, vol. 54, no. 1, pp. 107-113, 2007.
[30] R. J. Thomas, J. E. Mietus, C. K. Peng, and A. L. Goldberger, “An electrocardiogram-based technique to assess cardiopulmonary coupling during sleep,” Sleep, vol. 28, pp. 1151-61, 2005.
[31] S. Patil, H. Schneider, A. Schwartz and P. Smith, “Adult Obstructive Sleep Apnea: Pathophysiology and Diagnosis,” Chest, vol. 132, pp. 325-327, 2007.
[32] S. Chakravarthy and S. Kuo, “Application of Active Noise Control for Reducing Snore,” ICASSP Proceedings. IEEE International Conference, vol. 5, pp. 305-308, 2006.
[33] S. Choi, L. Bennett, R. Mullins, R. Davies and J. Stradling, “Which Derivative from Overnight Oximetry Best Predicts Symptomatic Response to Nasal Continuous Positive Airway Pressure in Patients with Obstructive Sleep Apnoea?,” Respiratory Medicine, vol. 94, no. 9, pp. 895-899, 2000.
[34] T. Young, M. Palta, J. Skatrud, S. Weber and S. Badr, “The Occurrence of Sleep-disordered Breathing Among Middle-aged Adults,” New England Journal of Medicine, vol. 328, pp. 1230-1235, 1993.
[35] U. Magalang, J. Dmochowski, S. Veeramachaneni, A. Draw, M. Mador, A. El-Solh and B. Grant, “Prediction of the Apnea-hypopnea Index From Overnight Pulse Oximetry,” Chest, vol. 124, no. 5, pp. 1694-1701, 2003.
[36] W. Fan, S. Stolfo, J. Zhang and P. Chan “Adacost: Misclassication Cost-sensitive Boosting,” In Proceedings of Sixteenth International Conference on Machine Learning, pp. 97-105, 1999.
[37] Z. Zhou and X. Liu, “On Multi-class Cost-sensitive Learning,” in Proceedings of the 21st National Conference on Artificial Intelligence, vol. 26, pp. 232-257, 2006.
[38] 王人鋒, “使用新基線定義與腦波清醒資訊改善血氧濃度特徵對呼吸暫止症的預測,” 中山大學博士學位論文, 2009.
[39] 陳崇賢, “失眠之評估與治療,” 家庭醫學與基層醫療 , 2010.
[40] 張可臻, “淺談阻塞性睡眠呼吸中止症的評估、診斷與治療,” 家庭醫學與基層醫療 , 2011.