本研究旨在使用心電圖之連續小波轉換時頻圖以及離散小波轉換作為特徵，結合深度學習演算法以及機械學習演算法，並搭配套用合成少數類過取樣技術(SMOTE)，用於辨識睡眠中的睡眠呼吸中止/低通氣量之演算法。本研究採用成大醫院睡眠中心所提供之資料庫以及MIT Physionet Sleep Apnea資料庫。在成大醫院所提供的睡眠多通道生理訊號中，本研究先以k-fold交叉驗證方法，當使用60秒為一判斷單位，在不使用以及使用SMOTE可分別得到94.7%以及95.5% 的準確率。當使用10秒為一判斷單位，在不使用以及使用SMOTE可分別得到84.9% and 93.8% 的準確率。接著使用Leave-one-subject-out交叉驗證方法，平均辨識準確率為82.3%。在MIT Physionet Sleep Apnea所提供的睡眠多通道生理訊號中，本研究比較了深度學習以及機械學習演算法在60秒的判斷單位下、判斷是否有呼吸睡眠中止，深度學習演算法達到 93.4%、機械學習得到85.2%的辨識率。Leave-one-subject-out交叉驗證方法的準確率則為83.1%.。本研究成功了使用套用合成少數類過取樣技術、心電圖訊號、深度學習/機械學習演算法達到辨識睡眠呼吸中止、睡眠低通氣量、睡眠正常呼吸的目標。

This thesis proposes a sleep apnea/hypopnea event classification algorithm using machine learning/deep learning approaches and time-frequency wavelet derived from ECG signal. Time-frequency transformation is the main method to perform signal processing. The wavelet transform includes continuous wavelet transform (CWT), and discrete wavelet transform (DWT) were utilized in this thesis to produce ECG spectrogram and decomposition signals. The thesis used signal preprocessing to normalize raw signal and remove noise and artifacts, the time-frequency transformation of ECG signal through CWT and DWT. Deep learning approach using ResNet-50 model to classify ECG spectrogram at different frequency bands. A DWT of ECG signal was used as feature extraction, which decomposes the original signal into a set of coefficients of level 10. Feature selection was also adopted to find an optimum number of features for the classification model. These features are input to the RBF-SVM classifier to detect sleep apnea, hypopnea, and normal events. The results were explained through per-segment and per-subject classification, then validated by k-fold cross-validation and leave-one-subject-out. The time length of the ECG signal was set as 10 and 60 seconds to identify the minimum time window length to achieve satisfactory performance.
Two databases from National Cheng Kung University Hospital and PhysioNet were used for verification. In NCKUH consisted of an imbalanced number of each event (apnea, hypopnea, and normal), so the SMOTE technique was applied to deal with this problem. In the 60s spectrogram, the accuracy of the original dataset and SMOTE dataset are 94.7% and 95.5% in the 0.5-50 Hz frequency band, respectively. In the 10s spectrogram, the accuracy of 80.4% in both balanced and imbalanced datasets in the 8-50 Hz band. The result of DWT was also increased by using SMOTE method in 60s segments with an accuracy of 84.9% and 93.8% in 10s segments. For per-subject classification, the average accuracy is 82.3%. The PhysioNet database used for identification apnea or without apnea. In the 60s segment, the accuracy of 93.4% can be obtained in the 0.8-10 Hz frequency band and 85.2% accuracy for DWT classification. In the 10s segment, the result can achieve 88.0% in 0.8-10 Hz frequency band and 93.4% accuracy in DWT. The per-subject classification was employed to obtain a good average accuracy of 83.1%. In the literature comparison, the contribution of this study is to discriminate particularly apnea, hypopnea, and normal breathing, which previous studies did not perform. The result can achieve better accuracy in one minute. Every subject’s assessment was considered a more realistic scheme for clinical application.

1. P. E. Peppard, T. Young, J. H. Barnet, et al., “Increased prevalence of sleep-disordered breathing in adults,” American journal of epidemiology, vol. 177, no. 9, pp. 1006-1014, 2013.
2. S. Quan, J. C. Gillin, M. Littner, et al., “Sleep-related breathing disorders in adults: Recommendations for syndrome definition and measurement techniques in clinical research," Sleep (New York, NY), vol. 22, no. 5, pp. 662-689, 1999.
3. J. Solà-Soler, B. F. Giraldo, J. A. Fiz, et al., “Relationship between heart rate excursion and apnea duration in patients with Obstructive Sleep Apnea,” 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). 2017.
4. “Clinical guideline for the evaluation, management and long-term care of obstructive sleep apnea in adults,” Journal of clinical sleep medicine, vol. 5, no. 3, pp. 263-276, 2009.
5. O. Kadarullah and Y. Annisa, “Pengaruh Obstructive Sleep Apnea (OSA) Terhadap Terjadinya Hipertensi Di Poli Saraf RSUD Prof. Dr. Margono Soekarjo,” Sainteks, vol. 13, no. 2, 2017.
6. B. Richard B, Q. Stuart F, A. Alexandre R, et al., “The AASM Manual for the Scoring of Sleep and Associated Events ver 2.6,” 2020.
7. J. C. Jun, S. Chopra, and A. R. Schwartz, “Sleep apnoea,” European Respiratory Review, vol. 25, no. 139, pp. 12-18, 2016.
8. N. Febriana, A. Rizal, and E. Susanto, “Sleep monitoring system based on body posture movement using Microsoft Kinect sensor,” in AIP Conference Proceedings, AIP Publishing LLC, 2019.
9. O. Elisha, A. Tarasiuk, and Y. Zigel, “Detection of Obstructive Sleep Apnoea Using Speech Signal Analysis,” Detection of Obstructive Sleep Apnoea Using Speech Signal Analysis., pp. 13-16, 2011.
10. Available from: https://www.canstockphoto.com/snoring-and-sleep-apnea-eps10-11687428.html.
11. L. S. Lilly, “Pathophysiology of heart disease: a collaborative project of medical students and faculty,” 2012: Lippincott Williams & Wilkins.
12. C. Varon, A. Caicedo, D. Testelmans, et al., “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.
13. C. Varon, J. Morales, J. Lázaro, et al., “A Comparative Study of ECG-derived Respiration in Ambulatory Monitoring using the Single-lead ECG,” Scientific reports, vol. 10, no. 1, pp. 1-14, 2020.
14. A. H. Khandoker, J. Gubbi, and M. Palaniswami, “Automated scoring of obstructive sleep apnea and hypopnea events using short-term electrocardiogram recordings,” IEEE Transactions on Information Technology in Biomedicine, vol. 13, no. 6, pp. 1057-1067, 2009.
15. T. Penzel, J. W. Kantelhardt, L. Grote, et al., “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.
16. A. Hachem, M. Ayache, L. El Khansa, et al., “ECG classification for Sleep Apnea detection,” 3rd Middle East Conference on Biomedical Engineering (MECBME), IEEE, 2016.
17. D. S. Sisodia, K. Sachdeva, and A. Anuragi, “Sleep order detection model using support vector machines and features extracted from brain ECG signals,” International Conference on Inventive Computing and Informatics (ICICI), IEEE, 2017.
18. F. d. r. Roche, J.-M. Gaspoz, I. Court-Fortune, et al., “Screening of obstructive sleep apnea syndrome by heart rate variability analysis,” Circulation, vol. 100, no. 13, pp. 1411-1415, 1999.
19. A. Prabha, A. Trivedi, A. A. Kumar, et al., “Automated system for obstructive sleep apnea detection using heart rate variability and respiratory rate variability,” International Conference on Advances in Computing, Communications and Informatics (ICACCI), IEEE, 2017.
20. A. Zarei and B. M. Asl, “Automatic detection of obstructive sleep apnea using wavelet transform and entropy-based features from single-lead ECG signal,” IEEE journal of biomedical and health informatics, vol. 23, no. 3, pp. 1011-1021, 2018.
21. U. R. Acharya, E. C.-P. Chua, O. Faust, et al., “Automated detection of sleep apnea from electrocardiogram signals using nonlinear parameters,” Physiological measurement, vol. 32, no. 3, pp. 287, 2011.
22. S. Shao, T. Wang, C. Song, et al., “Obstructive Sleep Apnea Recognition Based on Multi-Bands Spectral Entropy Analysis of Short-Time Heart Rate Variability,” Entropy, vol. 21, no. 8, pp. 812, 2019.
23. J.-U. Park, H.-K. Lee, J. Lee, et al., “Automatic classification of apnea/hypopnea events through sleep/wake states and severity of SDB from a pulse oximeter,” Physiological measurement, vol. 36, no. 9, pp. 2009, 2015.
24. U. Erdenebayar, Y. J. Kim, J. U. Park, et al., “Deep learning approaches for automatic detection of sleep apnea events from an electrocardiogram,” Computer methods and programs in biomedicine, vol. 180, pp. 105001, 2019.
25. R. Khincha, S. Krishnan, R. Parveen, et al., “ECG Signal Analysis on an Embedded Device for Sleep Apnea Detection,” in International Conference on Image and Signal Processin, 2020, Springer.
26. M. J. Rahman, R. Mahajan, and B. I. Morshed. “Severity classification of obstructive sleep apnea using only heart rate variability measures with an ensemble classifier,” IEEE EMBS International Conference on Biomedical & Health Informatics (BHI), 2018.
27. P. De Chazal and N. Sadr, “Sleep apnoea classification using heart rate variability, ECG derived respiration and cardiopulmonary coupling parameters,” 38th Annual international conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2016.
28. H. Sharma and K. Sharma, “An algorithm for sleep apnea detection from single-lead ECG using Hermite basis functions,” Computers in biology and medicine, vol. 77, pp. 116-124, 2016.
29. T. Wang, C. Lu, and G. Shen, “Detection of sleep apnea from single-lead ECG signal using a time window artificial neural network,” BioMed research international, vol. 2019, 2019.
30. S. A. Singh and S. Majumder, “A novel approach osa detection using single-lead ECG scalogram based on deep neural network,” Journal of Mechanics in Medicine and Biology, vol. 19, no. 04, pp. 1950026, 2019.
31. S. I. Niroshana, X. Zhu, K. Nakamura, et al., “A fused-image-based approach to detect obstructive sleep apnea using a single-lead ECG and a 2D convolutional neural network,” Plos one, vol. 16, no. 4, pp. e0250618, 2021.
32. H.-Y. Chang, C.-Y. Yeh, C.-T. Lee, et al., “A sleep apnea detection system based on a one-dimensional deep convolution neural network model using single-lead electrocardiogram,” Sensors, vol. 20, no. 15, pp. 4157, 2020.
33. G. Moody, R. Mark, A. Goldberger, et al., “Stimulating rapid research advances via focused competition: The Computers in Cardiology Challenge 2000,” Computers in Cardiology 2000, vol. 27 (Cat. 00CH37163), 2000, IEEE.
34. Available from: https://ataspinar.com/2018/12/21/a-guide-for-using-the-wavelet-transform-in-machine-learning/.
35. R. S. Pathak, “The wavelet transform,” Vol. 4. 2009: Springer Science & Business Media.
36. L.-Y. Shyu, Y.-H. Wu, and W. Hu, “Using wavelet transform and fuzzy neural network for VPC detection from the Holter ECG,” IEEE Transactions on Biomedical Engineering, vol. 51, no. 7, pp. 1269-1273, 2004.
37. I. R. Legarreta, P. S. Addison, M. Reed, et al., “Continuous wavelet transform modulus maxima analysis of the electrocardiogram: beat characterisation and beat-to-beat measurement,” International Journal of Wavelets, Multiresolution and Information Processing, vol. 3, no. 01, pp. 19-42, 2005.
38. M. A. Shams, H. I. Anis, and M. El-Shahat, “Denoising of Heavily Contaminated Partial Discharge Signals in High-Voltage Cables Using Maximal Overlap Discrete Wavelet Transform,” Energies, vol. 14, no. 20, pp. 6540, 2021.
39. K. N. Rajesh, R. Dhuli, and T. S. Kumar, “Obstructive sleep apnea detection using discrete wavelet transform-based statistical features,” Computers in Biology and Medicine, vol. 130, pp. 104199, 2021.
40. C. E. Shannon, “A mathematical theory of communication,” ACM SIGMOBILE mobile computing and communications review, vol. 5, no. 1, pp. 3-55, 2001.
41. R. F. Rushmer, “Cardiovascular dynamics,” Academic Medicine, vol. 36, no. 6, pp. 742, 1961.
42. T. S. Kumar and V. Kanhangad, “Detection of electrocardiographic changes in partial epileptic patients using local binary pattern based composite feature,” Australasian physical & engineering sciences in medicine, vol. 41, no. 1, pp. 209-216, 2018.
43. Available from: https://www.mathworks.com/help/stats/neighborhood-component-analysis.html.
44. J. Jiang, H. Zhang, D. Pi, et al., “A novel multi-module neural network system for imbalanced heartbeats classification,” Expert Systems with Applications: X, vol. 1, pp. 100003, 2019.
45. N. V. Chawla, K. W. Bowyer, L. O. Hall, et al., “SMOTE: synthetic minority over-sampling technique,” Journal of artificial intelligence research, vol. 16, pp. 321-357, 2002.
46. A. Vilamala, K. H. Madsen, and L. K. Hansen, “Deep convolutional neural networks for interpretable analysis of EEG sleep stage scoring,” 27th international workshop on machine learning for signal processing (MLSP), 2017, IEEE.
47. H. Yue, Y. Lin, Y. Wu, et al., “Deep Learning for Diagnosis and Classification of Obstructive Sleep Apnea: A Nasal Airflow-Based Multi-Resolution Residual Network,” Nature and Science of Sleep, vol. 13, pp. 361, 2021.
48. H. Yue, Y. Lin, Y. Wu, et al., “Deep Learning for Diagnosis and Classification of Obstructive Sleep Apnea: A Nasal Airflow-Based Multi-Resolution Residual Network,” Nature and Science of Sleep, vol. 13, pp. 361, 2021.
49. W. S. Noble, “What is a support vector machine?,” Nature biotechnology, vol. 24, no. 12, pp. 1565-1567, 2006.
50. K. Thurnhofer-Hemsi, E. López-Rubio, M. A. Molina-Cabello, et al., “Radial basis function kernel optimization for Support Vector Machine classifiers,” arXiv preprint arXiv:2007.08233, 2020.
51. Y. Jung, “Multiple predicting K-fold cross-validation for model selection,” Journal of Nonparametric Statistics, vol. 30, no. 1, pp. 197-215, 2018.
52. K. Liao, M. V. Mollison, T. Curran, et al., “Single-trial EEG predicts memory retrieval using leave-one-subject-out classification,” international conference on bioinformatics and biomedicine (bibm), 2018, IEEE.
53. Available from: https://towardsdatascience.com/accuracy-recall-precision-f-score-specificity-which-to-optimize-on-867d3f11124.
54. W. R. Ruehland, P. D. Rochford, F. J. O’Donoghue, et al., “The new AASM criteria for scoring hypopneas: impact on the apnea hypopnea index,” Sleep, vol. 32, no. 2, pp. 150-157, 2009.
55. G. Surrel, A. Aminifar, F. Rincón, et al., “Online obstructive sleep apnea detection on medical wearable sensors,” IEEE transactions on biomedical circuits and systems, vol. 12, no. 4, pp. 762-773, 2018.
56. G. Strang, “Wavelets and dilation equations: A brief introduction,” SIAM review, vol. 31, no. 4, pp. 614-627, 1989.
57. T. J. Burns, S. K. Rogers, D. W. Ruck, et al., “Discrete, spatiotemporal, wavelet multiresolution analysis method for computing optical flow,” Optical Engineering, vol. 33, no. 7, pp. 2236-2247, 1994.
58. A. Grossmann and J. Morlet, “Decomposition of Hardy functions into square integrable wavelets of constant shape,” SIAM journal on mathematical analysis, vol. 15, no. 4, pp. 723-736, 1984.
59. C.-y. Lin, Y.-W. Wang, F. Setiawan, et al., “Sleep Apnea Classification Algorithm Development Using a Machine-Learning Framework and Bag-of-Features Derived from Electrocardiogram Spectrograms,” Journal of Clinical Medicine, vol. 11, no. 1, pp. 192, 2021.