睡眠在生活中是非常重要的一個環節，人的一生中約有1/3 以上的時間處於
睡眠狀況下，而睡眠也是人體生理以及心理修復的黃金時段，但在現今生活節奏
緊張的時代，所帶來的壓力導致能夠正確且安穩睡眠的人是少之又少。不正確睡
眠所帶來的危害，可能遠比你想像中來的嚴重，舉凡憂鬱症、加速衰老、精神不
濟、失智等，各種慢性疾病都有可能伴隨而來。
為了讓生活忙碌、節奏緊張的現代人能夠擁有良好的睡眠品質，在科技進步
的現在已經有許多穿戴式裝置具有監控睡眠狀態，並且分析睡眠品質的功能，讓
使用者進一步的改善自己的睡眠狀況，但部分的裝置礙於硬體的設計無法植入太
多的感測器；而部分的裝置其檢測功能僅為產品推出的簡易附帶功能。在睡眠分
析的部分，多數的裝置皆是僅使用加速度計的三軸變化量來判別睡眠時期的體動
狀態，並沒有辦法非常準確的判斷出受試者的睡眠狀態，導致睡眠分析的準確率
不高。
因此本論文提出一套演算法，利用睡眠狀態的心電訊號（Electrocardiography,
ECG），進行心率變異度（Heart Rate Variability, HRV）的時域、頻域、非線性分
析以及由心電訊號提取的呼吸信號（ECG-Derived Respiration, EDR）等總共12
項特徵進行支持向量機（Support Vector Machine, SVM）的分類（Classify），進行
睡眠狀態的分析，將人體的睡眠狀態分類成清醒、淺眠以及深眠三種階段，整體
準確率在三種不同的訓練資料比例（20%、33%、60%）分別可以達到74.36%、
77.14%及84.64%。

Sleep plays a key part throughout human life. About one-third of human life is in sleep state, which is also a prime time for physiological and psychological repairs. However, modern people usually live a busy life at high pressure, and only few people can get a proper and sound sleep. The harm caused by not getting good sleep can be unimaginable and it may cause diseases such as depression, accelerated aging, mental disability, and dementia.
In order to help busy and fast-paced modern people get better sleep quality, nowadays many wearable devices have the function of monitoring sleep state and analyzing sleep quality for users to further improve their sleep quality. However, on some of the devices, only few sensors can be implanted due to limits of their hardware design, and some of the devices only equip with simple added detection functions. Since most wearable devices only use the acceleration signals to analyze the users’ body motions in sleep state, which cannot accurately determinate the users’ sleep state, the accuracy of the sleep analysis is not high.
Therefore, in this study, an algorithm is developed to automatically classify sleep stages by extracting 12 features of heart rate variability (HRV), including time domain, frequency domain, non-linear analysis, ECG-derived respiration (EDR), and so on, from Electrocardiogram (ECG) signals every 30 seconds and a multi-stage Support Vector Machine (SVM) is used to classify all sleep epochs into three sleep stages (wake, light sleep, deep sleep). The accuracy of our proposal in 3 different training data proportion (20%, 33%, 60%) is up to 74.36%, 77.14%, and 84.64%.

[01]G. Graci, "Pathogenesis and Management of Cancer-Related Insomnia," Patho-genesis and Management of Cancer-Related Insomnia, vol. 3, no. 5, pp. 349-360, Sep. 2005.
[02]魏銘賢, "居家睡眠品監測系統之研製," 國立台北科技大學, 2008.
[03]陳品汶, "以心電圖為基礎之即時睡眠品質評估系統," 國立陽明大學, 2010.
[04]A. M. Bianchi, M. O. Mendez, and S. Cerutti, "Processing of signals recorded through smart devices: Sleep-quality assessment," IEEE Transactions on Information Technology in Biomedicine, vol. 14, no. 3, pp. 741-747, May 2010.
[05]林秤毅, "穿戴式裝置之睡眠品質偵測演算法分析與評估," 國立台北科技大學, 2014.
[06]Y. Chen, X. Zhu, and W. Chen, "Automatic sleep staging based on ECG signals using hidden Markov models," 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 530-533, Aug. 2015.
[07]M. Adnane and Z. Jiang, "Automatic sleep-wake stages classifier based on ECG," ICROS-SICE International Joint Conference 2009, pp. 493-499, Aug. 2009.
[08]H. Werteni, S. Yacoub, and N. Ellouze, "An Automatic Sleep-Wake Classifier Using ECG Signals," IJCSI International Journal of Computer Science Issues, vol. 11, no. 1, pp. 84-93, Jul. 2014.
[09]N. Surantha, S. M. Isa, T. F. Lesmana, and I. M. A. Setiawan, "Sleep stage classi-fication using the combination of SVM and PSO," 2017 1st International Confer-ence on Informatics and Computational Sciences (ICICoS), pp. 177-182, Nov. 2017.
[10]E. Aserinsky and N. Kleitman, "Regularly occurring periods of eye motility, and concomitant phenomena, during sleep," Science, vol. 118, no. 3062, pp. 273-274, Sep. 1953.
[11]R. Allan and K. Anthony, "A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects. Allan Rechtschaffen and An-thony Kales, editors," U. S. National Institute of Neurological Diseases and Blindness, Neurological Information Network, pp. 246-247, Feb. 1968.
[12]K. Richards, C. Nagel, M. Markie, J. Elwell, and C. Barone, "Use of complemen-tary and alternative therapies to promote sleep in critically ill patients," Critical care nursing clinics of North America, vol. 15, no. 3, pp. 329-340, Sep. 2003.
[13]M. Hirshkowitz et al., "National Sleep Foundation’s sleep time duration recom-mendations: methodology and results summary," Sleep Health, vol. 1, no. 1, pp. 40-43, Mar. 2015.
[14]H. P. Roffwarg, J. N. Muzio, and W. C. Dement, "Ontogenetic development of the human sleep-dream cycle," Science, vol. 152, no. 3722, pp. 604-619, Apr. 1966.
[15]E. V. Cauter, R. Leproult, and L. Plat, "Age-related changes in slow wave sleep and REM sleep and relationship with growth hormone and cortisol levels in healthy men," Journal of the American Medical Association, vol. 284, no. 7, pp. 861-868, Aug. 2000.
[16]C. C. H. Yang, C. W. Lai, H. Y. Lai, and T. B. J. Kuo, "Relationship between electroencephalogram slow-wave magnitude and heart rate variability during sleep in humans," Neuroscience Letters, vol. 329, no. 2, pp. 213-216, Aug. 2002.
[17]J. M. Monti, F. Alvariño, and D. Monti, "Conventional and power spectrum anal-ysis of the effects of zolpidem on sleep EEG in patients with chronic primary in-somnia," Sleep, vol. 23, no. 8, pp. 1075-1084, Dec. 2000.
[18]S. A. Israel, ""Sleep is not tangible" or what the Hebrew tradition has to say about sleep," Psychosomatic Medicine, vol. 63, no. 5, pp. 778-787, Sep. 2001.
[19]E. Pedersen and K. P. Waye, "Wind turbine noise, annoyance and self-reported health and well-being in different living environments," Occupational and Envi-ronmental Medicine, vol. 64, no. 7, pp. 480-486, Jul. 2007.
[20]G. Bluhm, E. Nordling, and N. Berglind, "Road traffic noise and annoyance - an increasing environmental health problem," Noise & Health, vol. 6, no. 24, pp. 43-49, Jul. 2004.
[21]K. Morgan and S. J. Closs, "Causes and correlates of disturbed sleep," Sleep management in nursing practice: an evidence-based guide, pp. 39-60, Jul. 1999.
[22]B. A. Hilton, "Quantity and quality of patients’ sleep and sleep-disturbing factors in a respiratory intensive care unit," Journal of advanced nursing, vol. 1, no. 6, pp. 453-468, Nov. 1976.
[23]P. M. Becker and A. O. Jamieson, "Common sleep disorders in the elderly: diag-nosis and treatment," Geriatrics, vol. 47, no. 3, pp. 41-52, Mar. 1992.
[24]T. Kendzerska, M. Kamra, B. J. Murray, and M. I. Boulos, "Incident cardiovascu-lar events and death in individuals with restless legs syndrome or periodic limb movements in sleep: A systematic review," Sleep, vol. 40, no. 3, Mar. 2017.
[25]R. K. Bogan, "Effects of restless legs syndrome (RLS) on sleep," Neuropsychiat-ric Disease and Treatment, vol. 2, no. 4, pp. 513-519, Dec. 2006.
[26]H. Kawamura and C. H. Sawyer, "Elevation in brain temperature during paradox-ical sleep," Science, vol. 150, no. 3698, pp. 912-913, Nov. 1965.
[27]眼裡是另一番美景, "什麼樣的心理因素會導致失眠的發生, "每日頭條. [Online]. Available: https://kknews.cc/zh-tw/health/3jrk4y3.html. [Accessed: 06-Nov-2018].
[28]A. Kales et al., "Biopsychobehavioral correlates of insomnia, III: Polygraphic findings of sleep difficulty and their relationship to psychopathology," International Journal of Neuroscience, vol. 23, no. 1, pp. 43-55, Mar. 1984.
[29]W. B. Mendelson, "Sleep after forty," American Family Physician, vol. 29, no. 1, pp. 135-139, 1984.
[30]K. Morgan, H. Dallosso, S. Ebrahim, T. Arie, and P. H. Fentem, "Characteristics of subjective insomnia in the elderly living at home," Age and ageing, vol. 17, no. 1, pp. 1-7, Jan. 1988.
[31]C. A. Kushida, B. M. Bergmann, and A. Rechtschaffen, "Sleep deprivation in the rat: IV. Paradoxical sleep deprivation," Sleep, vol. 12, no. 1, pp. 22-30, Feb. 1989.
[32]Robert Stickgold, "該睡覺了！, " 科學人雜誌, 2017. [Online]. Available: http://sa.ylib.com/MagArticle.aspx?Unit=featurearticles&id=3797. [Accessed: 06-Nov-2018].
[33]李宇宙, "身心疾病之睡眠障礙," 台灣醫學, 2000.
[34]G. Brandenberger, J. Ehrhart, and M. Buchheit, "Sleep stage 2: An electroenceph-alographic, autonomic, and hormonal duality," Sleep, vol. 28, no. 12, pp. 1535-1540, Dec. 2005.
[35]K. Spiegelhalder et al., "Heart rate and heart rate variability in subjectively report-ed insomnia," Journal of Sleep Research, vol. 20, no. 1, pp. 137-145, Mar. 2011.
[36]M. H. Bonnet and D. L. Arand, "Heart rate variability in insomniacs and matched normal sleepers," Psychosomatic Medicine, vol. 60, no. 5, pp. 610-615, Sep. 1998.
[37]M. H. Bonnet and D. L. Arand, "Heart rate variability: Sleep stage, time of night, and arousal influences," Electroencephalography and Clinical Neurophysiology, vol. 102, no. 5, pp. 390-396, Jun. 1998.
[38]G. B. Edwards and L. M. Schuring, "Pilot study: validating staff nurses’ observa-tions of sleep and wake states among critically ill patients, using polysomnogra-phy," American journal of critical care: an official publication, American Associa-tion of Critical-Care Nurses, vol. 2, no. 2, pp. 125-32, Apr. 1994.
[39]R. Lerner, "Sleep loss in the aged: implications for nursing practice," Journal of Gerontological Nursing, vol. 8, no. 6, pp. 323-326, Jun. 1982.
[40]D. Cohen et al., "Sleep Disturbances in the Institutionalized Aged," Journal of the American Geriatrics Society, vol. 31, no. 2, pp. 79-82, Feb. 1983.
[41]C. R. Soldatos, D. G. Dikeos and, T. J. Paparrigopoulos, "Athens Insomnia Scale: Validation of an instrument based on ICD-10 criteria," Journal of Psychosomatic Research, vol. 48, no. 6, pp. 555-560, Jun. 2000.
[42]M. W. Johns, "A new method for measuring daytime sleepiness: the Epworth sleepiness scale," Sleep, vol. 14, no. 6, pp. 540-545, Dec. 1991.
[43]C. M. Morin, "Insomnia: Psychological assessment and management," New York, NY, US: Guilford Press, May 1993.
[44]E. Hoddes, V. Zarcone, H. Smythe, R. Phillips, and W. C. Dement, "Quantifica-tion of Sleepiness: A New Approach," Psychophysiology, vol. 10, no. 4, pp. 431-436, Jul. 1973.
[45]S. S. Kazarian, M. G. Howe and, K. G. Csapo, "Development of the sleep be-havior self-rating scale," Behavior Therapy, vol. 10, no. 3, pp. 412-417, Jun. 1979.
[46]M. Ohayon et al., "National Sleep Foundation’s sleep quality recommendations: first report," Sleep Health, vol. 3, no. 1, pp. 6-19, Feb. 2017.
[47]D. J. Buysse, C. F. Reynolds, T. H. Monk, S. R. Berman, and D. J. Kupfer, "The Pittsburgh sleep quality index: A new instrument for psychiatric practice and re-search," Psychiatry Research, vol. 28, no. 2, pp. 193-213, May 1989.
[48]G. H. Klem, H. O. Lüders, H. H. Jasper and, C. Elger, "The ten-twenty electrode system of the International Federation. The International Federation of Clinical Neurophysiology," Electroencephalography and clinical neurophysiology. Sup-plement, vol. 52, pp. 3-6, 1999.
[49]"(2002-07-20) 睡眠障礙與睡眠評估 - Optimal Design Lab, YZU. " [Online]. Available: http://designer.mech.yzu.edu.tw/articlesystem/article/compressedfile/(2002-07-20) 睡眠障礙與睡眠評估.aspx?ArchID=444. [Accessed: 23-Jan-2019].
[50]S. S. Barold, "Willem Einthoven and the birth of clinical electrocardiography a hundred years ago," Cardiac Electrophysiology Review, vol. 7, no. 1, pp. 99-104, Jan. 2003.
[51]J. Burstein and L. Ellenbogen, "Electrocardiograms in which the main initial ven-tricular deflections are directed downward in the standard leads," American Heart Journal, vol. 16, no. 2, pp. 165-174, Aug. 1938.
[52]F. N. Wilson, F. D. Johnston, A. G. Macleod and, P. S. Barker, "Electrocardio-grams that represent the potential variations of a single electrode," American Heart Journal, vol. 9, no. 4, pp. 447-458, Apr. 1934.
[53]J. Kim and E. André, "Emotion recognition based on physiological changes in music listening," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 30, no. 12, pp. 2067-2083, Feb. 2008.
[54]L. E. C. Martínez et al., "Maximum Heart Rate during exercise: Reliability of the 220-age and Tanaka formulas in healthy young people at a moderate elevation," Revista de la Facultad de Medicina, Universidad Nacional de Colombia, vol. 62, no. 4, pp. 579-585, May 2015.
[55]U. R. Acharya, K. P. Joseph, N. Kannathal, C. M. Lim, and J. S. Suri, "Heart rate variability: A review," Medical and Biological Engineering and Computing, vol. 44, no. 12, pp. 1031-1051, Dec. 2006.
[56]翁根本, 何慈育, 歐善福, 林竹川, 謝凱生, "心律變動性分析," 台灣醫界, vol. 52, no. 6, pp. 12-15, Dec. 2009.
[57]M. Malik, "Heart rate variability: Standards of measurement physiological inter-pretation, and clinical use," Circulation, vol. 93, no. 5, pp. 1043-1065, Mar. 1996.
[58]D. Longo et al., "Harrison’s Principles of Internal Medicine, 18th edition," McGraw-Hill, USA, 2012.
[59]D. Widjaja et al., "ECG-derived respiration: Comparison and new measures for respiratory variability," 2010 Computing in Cardiology, pp. 149-152, Sep. 2010.
[60]J. Boyle, N. Bidargaddi, A. Sarela, and M. Karunanithi, "Automatic detection of respiration rate from ambulatory single-lead ECG," IEEE Transactions on Infor-mation Technology in Biomedicine, vol. 13, no. 6, pp. 890-896, Nov. 2009.
[61]F. Liao, F. Wang, and P. Zhou, "Deriving respiratory signals from ECG by filter-ing method," 2011 3rd International Conference on Computer Research and De-velopment, pp. 456-459, Mar. 2011.
[62]S. B. Park, Y. S. Noh, S. J. Park, and H. R. Yoon, "An improved algorithm for respiration signal extraction from electrocardiogram measured by conductive tex-tile electrodes using instantaneous frequency estimation," Medical and Biological Engineering and Computing, vol. 46, no. 2, pp. 147-158, Feb. 2008.
[63]J. Lázaro et al., "Electrocardiogram Derived Respiratory Rate from QRS Slopes and R-Wave Angle," Annals of Biomedical Engineering, vol. 42, no. 10, pp. 2072-2083, Oct. 2014.
[64]"PhysioNet," [Online]. Available: https://www.physionet.org/. [Accessed: 26-Dec-2018].
[65]J. Pan and W. J. Tompkins, "A Real-Time QRS Detection Algorithm," IEEE Transactions on Biomedical Engineering, vol. BME-32, no. 3, pp. 230-263, Mar. 1985.
[66]"Complete Pan Tompkins Implementation ECG QRS detector - File Exchange - MATLAB Central. " [Online]. Available:https://ww2.mathworks.cn/matlabcentral/fileexchange/45840-complete-pan-tompkins-implementation-ecg-qrs-detector. [Accessed: 23-Jan-2019].
[67]Jramshur, "freqDomainHRV.m. " [Online]. Available:https://github.com/jramshur/HRVAS/blob/master/freqDomainHRV.m. [Accessed: 8-Feb-2019].
[68]J. Piskorski and P. Guzik, "Filtering Poincaré plots," Computational Methods in Science and Technology, vol. 11, no. 1, pp. 39-48, Dec. 2005.
[69]A. W. Whitney, "A Direct Method of Nonparametric Measurement Selection," IEEE Transactions on Computers, vol. C-20, no. 9, pp. 1100-1103, Sep. 1971.
[70]C. Cortes and V. Vapnik, "Support-Vector Networks," Machine Learning, vol. 20, no. 3, pp. 273-297, Sep. 1995.
[71]"HRVAS: HRV Analysis Software download | SourceForge.net. " [Online]. Available: https://sourceforge.net/projects/hrvas/. [Accessed: 8-Feb-2019].
[72]E. A. P. J. Prawiro, C. I. Yeh, N. K. Chou, M. W. Lee, and Y. H. Lin, "Integrated Wearable System for Monitoring Heart Rate and Step during Physical Activity," Mobile Information Systems, vol. 2016, pp. 1-10, Apr. 2016.