「睡眠呼吸中止症」為一種較常見的睡眠障礙，其症狀是在睡眠期間，會發生呼吸暫停或是呼吸緩慢的情況，造成患者較易發生低血氧以及睡眠中斷，所以患者通常睡眠品質較差，發生心血管疾病的機率也會比一般人高。多項研究報告指出，呼吸時胸腔體積的變化，會使心電訊號的振幅大小改變，且呼吸速率的快慢也與心率變化有著高度相關。本研究將會透過PhysioNet之MIT-BIH Polysomnographic Database資料庫所提供的心電訊號，利用由R波時距所形成的心源性呼吸訊號 (ECG-Derived Respiratory signal, EDR signal) 來觀察睡眠時的呼吸狀況。我們觀察到，當在呼吸中止發生前或發生的前半期間，心源性呼吸訊號會上升，而在呼吸中止的後段或結束時，心源性呼吸訊號會緩慢下降，因此，我們將會以此走向特徵來偵測是否有睡眠呼吸中止症的發生。本論文為了偵測睡眠呼吸中止症，我們提出一種分為兩階段的偵測方法，首先會找出所有潛在且符合特性的走向，在第二階段會盡可能地排除掉false positive的部分，最後會與資料庫中所提供的呼吸狀況做比較，得出本方法的成效與可行性。

Sleep apnea is a common sleep disorder. Symptoms of sleep apnea include apnea and hypopnea happening during sleep, those cause patients to have lower blood oxygen level and disruptions to the normal sleep-wake cycle. According to above, people who get sleep apnea, usually have worse sleep quality and have higher probability to face cardiovascular disease. Numerous papers indicate that changes in chest volume during breathing will vary amplitudes of the ECG signal. Besides, the high correlation between the respiratory rate and heart rate is testified. In our work, we use ECG-derived respiratory signal (EDR signal), which is formed with RR interval time series through the ECG signal provided by the MIT-BIH Polysomnographic Database on PhysioNet, to observe the breathing condition during sleep. We notice that EDR signal rises before or at the beginning of apnea, then, it slowly drops during apnea or after apnea ends. Therefore, we will use the feature of trends to classify whether sleep apnea happens or not. In this thesis, in order to detect sleep apnea, we propose two-phase detection method. In first stage, it may find out all potential and qualified trends. In next stage, it may filter out false positives as many as possible. Finally, we will compare the outcome with annotations offered by database, concluding the performance and feasibility of this method.

致謝 I
中文摘要 II
Abstract III
Contents IV
List of Figures VI
List of Tables VIII
Chapter 1 Introduction 1
1.1 Research Motivation and Purposes 1
1.2 Research Method 2
1.3 Thesis Architecture 2
Chapter 2 General Background Information 3
2.1 Introduction to Electrocardiography 3
2.1.1 Fundamental of Electrocardiography 3
2.1.2 Cardiac Action Potential 3
2.1.3 Cardiac Conduction System 4
2.1.4 Measurement of ECG 5
2.1.5 ECG Waveform 8
2.2 Introduction to ECG-Derived Respiratory Signal 10
2.2.1 Traditional Detection of Sleep-Related Disorders 10
2.2.2 Fundamental of ECG-Derived Respiratory Signal 11
2.2.3 Relationship Between Respiration and Axis of Mean QRS Vector 12
2.2.4 Information About Respiration from ECG Signal 13
2.2.5 Methods for ECG-Derived Respiratory Signal 13
2.2.5.1 QRS Complex Area Ratio of Two Leads 13
2.2.5.2 R Wave Amplitude 14
2.2.5.3 R-to-S Wave Amplitude 15
2.2.5.4 RR Interval and Instantaneous Heart Rate 15
2.3 Introduction of Sleep Apnea 16
2.3.1 Definition of Hypopnea 17
2.3.2 Definition of Obstructive Sleep Apnea (OSA) 18
2.3.3 Definition of Central Sleep Apnea (CSA) 18
2.3.3 Definition of Apnea-Hypopnea-Index (AHI) 20
2.3.4 The Relationship Between Spectrum of HRV and Apnea 21
2.4 Related Works About EDR Signal and Apnea Detection 22
2.4.1 Similarity Between EDR signal and Actual Respiratory Signal 22
2.4.2 Respiratory Rate from EDR Signals 23
2.4.3 Apnea Detection 24
2.5 Treatment for Apnea 27
2.6 Other Diseases Diagnosable by ECG Signal 28
2.7 Introduction to Performance Evaluation 29
2.8 Introduction to PhysioNet and MIT-BIH Polysomnographic Database 30
2.8.1 PhysioNet 30
2.8.2 MIT-BIH Polysomnographic Database 31
2.8.3 Analyzing Tools of PhysioNet 34
Chapter 3 Proposed Apnea Detection Algorithm 35
3.1 QRS Complex Detection 36
3.2 EDR Signal Acquisition 38
3.3 Apnea Trends on EDR Signal 39
3.4 Algorithm of Apnea Detection 40
3.4.1 Phase Ⅰ of Detection 41
3.4.2 Phase Ⅱ of Detection 50
Chapter 4 Results and Discussion 51
4.1 Simulation Results 51
4.1.1 Results of Phase Ⅰ 51
4.1.2 Sections of FN and FP After Phase Ⅰ 53
4.1.3 Results of Phase Ⅱ 55
4.1.4 Sections of FN After Phase Ⅱ 57
4.2 Adjustment to Considered Inconsistent Annotations 60
Chapter 5 Conclusion 66
References 67

[1] 鄭智銘,“睡眠障礙與睡眠評估,”元智大學老人福祉科技研究中心,7月,2002

[2] 馬素華, “阻塞型睡眠呼吸中止症候群簡介,”臺東大學人文學報,
第一卷第二期,2011, pp. 119-136

[3] George B. Moody, Roger G. Mark, Andrea Zoccola, and Sara Mantero, “Derivation of Respiratory Signals from Multi-lead ECGs,” Computers in Cardiology, 1985

[4] MICHAEL R. ROSEN MD, THAI PHAM PhD, “Principles of Gender-Specific Medicine.” 2004

[5] J. GORDON BETTS, PETER DESAIX et al., “Anatomy & Physiology,” 2013

[6] Wikipedia contributors, “Cardiac conduction system,” Wikipedia, https://en.wikipedia.org/wiki/Cardiac_conduction_system

[7] Wikipedia contributors, “Electrocardiography,” Wikipedia, https://en.wikipedia.org/wiki/Electrocardiography

[8] “12 Lead ECG Placement Guide,” Cables & Sensors https://www.cablesandsensors.com/pages/12-lead-ecg-placement-guide-with-illustrations#2

[9] LibreTexts, “Anatomy and Physiology (Boundless),” Anatomy and Physiology (Boundless) - Medicine LibreTexts

[10] 蕭子健, 謝景旭,“呼吸量測裝置之簡介,”科儀新知, 213期,12
月,2017

[11] Wikipedia contributors, “多項生理睡眠檢查,” Wikipedia, https://zh.wikipedia.org/wiki/%E5%A4%9A%E9%A0%85%E7%94%9F%E7%90%86%E7%9D%A1%E7%9C%A0%E6%AA%A2%E6%9F%A5

[12]陳彥文, “認識「橫膈膜式呼吸」,” 中榮醫訊, 231期, pp.18-20,

[13] Lingeng Zhao, “Derivation of respiration from electrocardiogram during heart rate variability studies,” Theses, Science in Biomedical Engineering, New Jersey Institute of Technology, May, 1994

[14] BIOPAC,” QRS Amplitude & Respiratory Modulation,”
https://www.biopac.com/knowledge-base/qrs-amplitiude-respiratory-modulation/

[15] Ramakrishnan A. G., Adarsh A., “R-wave Amplitude Changes and Atypical Heart Rate Changes Accompanying Breath Hold During Low Breathing Rates,” IEEE,2020

[16] Saeed Babaeizadeh, Sophia H. Zhou, Stephen D. Pittman, David P. White, “Electrocardiogram-derived respiration in screening of sleep-disordered breathing,” Journal of Electrocardiology, 2011

[17] Haipeng Liu, John Allen, Dingchang Zheng, and Fei Chen, “ Recent development of respiratory rate measurement technologies,” Institute of Physics and Engineering in Medicine, 2019

[18] Matthew L. Ho, Steven D. Brass, “Obstructive sleep apnea,” Neurology International, vol. 3, 2011

[19] Michael Sateia, “International Classification of Sleep Disorders,” American Academy of Sleep Medicine, 3th edition, 2014

[20] Chien-Chang Hsu, Ping-Ta Shih, “A novel sleep apnea detection system in electroencephalogram using frequency variation,” Expert Systems with Applications, 2011, pp. 6014-6024

[21] Wikipedia contributors, “呼吸暫停低通氣指數,” Wikipedia, https://zh.wikipedia.org/zh-tw/%E5%91%BC%E5%90%B8%E6%9A%82%E5%81%9C%E4%BD%8E%E9%80%9A%E6%B0%94%E6%8C%87%E6%95%B0

[22] Task Force of The European Society of Cardiology and The North American
Society of Pacing and Electrophysiology, “Heart rate variability. Standards of measurement, physiological interpretation, and clinical use,” American Heart Association Inc., European Society of Cardiology, 1996

[23] Saeed Babaeizadeh, David P. White, Stephen D. Pittman, Sophia H. Zhou, “Automatic detection and quantification of sleep apnea using heart rate variability,” Journal of Electrocardiology 43, 2010

[24] Lorena S Correa, Eric Laciar, Abel Torres, Raimon Jane, “Performance
evaluation of three methods for respiratory signal estimation from the
electrocardiogram,” IEEE, 2008

[25] Surita Sarkar, Saptak Bhattacherjee, Saurabh Pal, “Extraction of Respiration Signal From ECG For Respiratory Rate Estimation,” Michael Faraday IET International Summit, 2015

[26] A Sobron, I Romero, T Lopetegi, “Evaluation of Methods for Estimation of Respiratory Frequency from the ECG,” Computing in Cardiology, 2010

[27] Lorena S Correa, Eric Laciar, Vicente Mut, Abel Torres, Raimon Jané, “Sleep Apnea Detection based on Spectral Analysis of Three ECG - Derived Respiratory Signals,” IEEE, September, 2009

[28] Shu-Han Fan, Chia-Ching Chou, Wei-Chen Chen, and Wai-Chi Fang, “Real-Time Obstructive Sleep Apnea Detection from Frequency Analysis of EDR and HRV using Lomb Periodogram,” IEEE, 2015

[29] Pauline Guyot, Bruno Chenuel, El-Hadi Djermoune, Thierry Bastogne, “Early detection of Cheyne-Stokes breathing via ECG-derived respiration in patients with severe heart failure: a pilot study,” 45th Computing in Cardiology Conference, 2018

[30]鄒永恩, “睡眠呼吸中止症的預防與治療,” 中國醫藥大學附設醫院, 科室文章, May, 2013, https://www.cmuh.cmu.edu.tw/NewsInfo/NewsArticle?no=3176

[31]吳詠霓, “什麼是睡眠呼吸中止症？誰是高危險群？打呼如何改善？症狀、預防、治療懶人包,” 健康2.0, June, 2023, https://health.tvbs.com.tw/medical/340447

[32] NADHIPURAM V. Bhagavan et al., “Evaluation of Human Serum Albumin Cobalt Binding Assay for the Assessment of Myocardial Ischemia and Myocardial Infarction,” Clinical Chemistry, No. 4, 2003, p 581–585

[33] Rachita, “Differences Between Myocardial Ischemia and Myocardial Infarction,” http://www.differencebetween.net/science/health/differences-between-myocardial-ischemia-and-myocardial-infarction/

[34] M A Hasan, D Abbott and M Baumert, “Beat-to-beat QT interval variability and T-wave amplitude in patients with myocardial infarction,” Physiological Measurement.34, Published 19 August, 2013, pp. 1075-1083

[35] Wikipedia contributors, “Sensitivity and specificity,” Wikipedia, https://en.wikipedia.org/wiki/Sensitivity_and_specificity

[36] researchers and engineers at the MIT Laboratory for Computational Physiology and Beth Israel Deaconess Medical Center, “Brief Introduction,” PhysioNet, https://physionet.org/about/

[37] George Moody, “MIT-BIH Polysomnographic Database,” PhysioNet, August, 1999, https://physionet.org/content/slpdb/1.0.0/

[38] Ichimaru Y, Moody GB, “MIT-BIH Polysomnographic Database_ Signals and Annotations, ” PhysioNet, August, 1999, https://physionet.org/files/slpdb/1.0.0/slpdb.html

[39] PhysioBank ATM, https://archive.physionet.org/cgi-bin/atm/ATM

[40] Ary L. Goldberger et al., “PhysioBank, Physio Toolkit, and PhysioNet: Components of a new research resource for complex physiologic signals,” Circulation, vol. 101, no. 23, Jun. 2000

[41] Chandra Wijaya, Andrian, Mawaddah Harahap, Christnatalis, Mardi Turnip,
Arjon Turnip, “Abnormalities State Detection from P-Wave, QRS Complex,
and T-Wave in Noisy ECG,” Journal of Physics, 2019

[42] Carolina Varon et al., “A Comparative Study of ECG derived Respiration in Ambulatory Monitoring using the Single-lead ECG,” Scientific Reports, Nature Research, 2020

[43]張宗毅, “利用心電圖訊號波型特徵實現基於Android智慧型裝置之心肌缺血偵測系統,” 論文, 國立清華大學, January, 2019

[44]黃教暐, “利用心電圖訊號波型特徵實現基於Android智慧型裝置之心律不整偵測系統,” 論文, 國立清華大學, January, 2019

[45] Think-Leader, ltd, “心跳與心電圖簡介,” https://www.q5.com.tw/Info-center/Help/Why-ECG