本篇論文研究的主要目的是利用配戴在手腕及腳踝的慣性感測器所收集的慣性訊號，開發辨識整夜睡眠姿勢的演算法，並基於睡眠姿勢的變化及睡眠中活動量的特徵建立睡眠品質分析模型估測主觀之睡眠品質。
其中辨識的四個主要睡眠姿勢為: 平躺、左側躺、右側躺及趴睡。在不同的睡眠姿勢下重力加速度值在感測器各軸上的分量必不相同。然而，每個人擁有其個別的睡眠習慣與姿勢，因此各個姿勢的重力加速度值的範圍會因人而異產生變化。因此利用門檻值的方法對所有人的睡眠姿勢作分類會導致許多錯誤的辨識結果，然而，每個人相同的睡眠姿勢其重力加速度之讀值仍會有群聚的現象，因此不同的睡眠姿勢即會歸屬到不同的群集。基於上述考量，本研究使用層級式集群法( hierarchical agglomerative clustering, HAC)來實現睡眠姿勢分析。層級式集群法會依據每個人在不同睡姿下重力加速度的群聚值分析睡眠姿勢。因此，受個體睡眠姿勢的變異而產生的誤判可被有效的減少。本論文所提出的睡眠姿勢分析演算法經過驗證，可達92.26%的正確率。
基於上述之睡眠姿勢辨識演算法及依加速度值所轉換的活動量特徵，本論文亦建立睡眠品質分析模型來估測主觀之睡眠品質。以中文化維辛式睡眠品質量表(Verran and Snyder-Halpern sleep scale in Chinese version)作為個人主觀睡眠品質之評估標準，並針對中文化維辛式量表中「睡眠受擾」與「有效睡眠」二個面向之睡眠品質分別建構分析模型。此模型經過驗證，對於有效睡眠與睡眠不受干擾二個面向之睡眠品質分析之正確率分別可達61.4%與83.03%。
本論文結合睡眠姿勢辨識演算法與睡眠中活動量之特徵，完成一適用於日常睡眠之睡眠品質評估模型，並與主觀量表進行驗證，結果顯示本研究，提供了一個可估測主觀睡眠品質的可靠工具。

This thesis presents sleep position recognition algorithm and its analysis based on an acceleration worn on ankle. There are 4 major sleep positions: supine, left lateral, right lateral and prone. Since the body postures are different in different sleep positions, the acceleration values for the four positions are different. However, as different people have different sleep posture habits, the range of gravity values varies from person to person, which causes that the gravity values of different sleep positions in different people overlap slightly. Thus it is impossible to use a threshold for classifying the sleep positions from all the people. Despite of such a fact, for the same person the gravity values of the same positions are pretty much the same, while the gravity values in the different positions separate. Based on this consideration, this study develops a hierarchical agglomerative clustering (HAC)-based approach for sleep position analysis. The HAC clusters gravities of each person and perform sleep position estimation based on the clustering analysis. Thus the misclassification of some individual data points caused by the variation can be reduced. The proposed sleep position analysis algorithm can achieve an accuracy of 92.26 %.
Based on the results of the above sleep position recognition algorithm and activity characteristics computed from the acceleration values, this research also established analysis models to predict subjective sleep quality. The Verran and Snyder-Halpern sleep scale (VHS) in Chinese version is selected as the standards to assess the subjective sleep quality. The factors of "sleep disturbance" and "effective sleep" in VHS are selected as the desired target to construct sleep quality analysis models. The correction rates of the factors: "sleep disturbance" and "effective sleep" of the sleep quality analysis are 61.4% and 83.03% respectively.
The proposed approach combines sleep position recognition algorithms and the characteristics of sleep activity to complete a daily sleep quality assessment model, and the validation results show that the proposed approach provides a reliable tool to estimate the daily subjective sleep quality.

[1]K. Spiegel, E. Tasaii, R. Leproult, and E. V. Cauter , “Effects of poor and short sleep on glucose metabolism and obesity risk,” Nature Reviews Dndocrinology, vol. 5, pp. 253-261, 2009.
[2]F. P. Cappuccio, F. L. D’Elia, M. Strazzullo, M.D. Michelle, A. Miller, “Quantity and quality of sleep and incidence of type 2 diabetes: a systematic review and meta-analysis,” Diabetes Care, vol. 33, pp. 414-420, 2010.
[3]S. R. Patel and F. B. Hu, “Short sleep duration and weight gain: a systematic review,” Obesity (Silver Spring), vol.16, pp. 643–653, 2008.
[4]C. M. Baldwin, A. M. Ervin, M. Z. Mays, J. Robbins, S. Shafazand, J. Walsleben, T. Weaver, “Sleep Disturbances, Quality of Life, and Ethnicity: The Sleep Heart Health Study,” Journal of clinical sleep medicine, vol.6, no.2, pp.176-183,2010.
[5]Javaheri S, Storfer-Isser A, Rosen CL, Redline S, “Sleep Quality and Elevated Blood Pressure in Adolescents,” Circulation, vol. 118, pp. 1034-1040, 2008.
[6]F. P. Cappuccio, D. Cooper, L. D’Elia, P. Strazzullo, and M. A. Miller, “Sleep duration predicts cardiovascular outcomes: a systematic review and meta-analysis of prospective studies,” European Heart Journal, vol.32, no.12, pp. 1484-1492, 2011.
[7]Online available: http://sensewear.bodymedia.com/
[8]Online available: http://www.theactigraph.com/
[9]Online available: http://www.bmedical.com.au/shop/actiwatch.htm
[10]J. De Koninck, P. Gagnon, and S. Lallier, “Sleep positions in the young adult and their relationship with the subjective quality of sleep,” Sleep, vol. 6, no. 1, pp. 52-59, 1983.
[11]T. Åkerstedt, K. Hume, D. Minors, J. Waterhouse, “The meaning of good sleep: a longitudinal study of polysomnography and subjective sleep quality,” Journal of Sleep Research, vol. 3, no. 3, pp. 152-158, 1994.
[12]R. R. Rosa, and M. H. Bonnet, “Reported chronic insomnia is independent of poor sleep as measured by electroencephalography,” Psychosomatic Medicine, vol. 62, no. 4, pp. 474-482, 2000.
[13]D. J. Buysse, C. F. Reynolds, T. H. Monk, S. R. Berman, D. J. Kupfer, “The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research,” Psychiatry Research, vol. 28, no. 2, pp. 193-213, 1989.
[14]A. M. Swartz, S. J. Strath, D. R. Bassett, W. L. O’brien, G. A. King, and B. E. Ainsworth, “Objective sleep measures and subjective sleep satisfaction: how do older adults with insomnia define a good night’s sleep,” Psychology and Aging, vol. 13, no. 1, pp. 159-163, 1998.
[15]G. Keklund, T. Akerstedt, “Objective components of individual differences in subjective sleep quality,” Journal of Sleep Research, vol. 6, no. 4, pp. 217-220, 1997.
[16]A. Sadeh, P. J. Hauri, D. F. Kripke, P. Lavie,“The role of actigraphy in the evaluation of sleep disorders,” Sleep, vol. 18, no. 4, pp. 228-302, 1995.
[17]A. Vallières, C. M. Morin, “Actigraphy in the Assessment of Insomnia,” Sleep, vol. 26, no. 7, pp. 902-906, 2003.
[18]S. S. Tworoger, S. Davis, M. V. Vitiello, M. J. Lentz and A. McTiernan, “Factors associated with objective (actigraphic) and subjective sleep quality in young adult women,” Journal of Psychosomatic Research, vol. 59, no. 1, pp. 11-19, 2005.
[19]J. F. Van Den Berg, F. J. Van Rooij, H. Vos, J. H. Tulen, A. Hofman, H. M. Miedema, A. K. Neven, H. Tiemeier, “Disagreement between subjective and actigraphic measures of sleep duration in a population-based study of elderly persons,” Jounrnal of Sleep Research, vol. 17, no. 3, pp. 295-302, 2008.
[20]R. J. Cole, D. F. Kripke, W. Gruen, D. J. Mullaney, J. C. Gillin, “Automatic sleep/wake identification from wrist activity,” Sleep, vol. 15, no. 5, pp. 461-469, 1992.
[21]S. T. Aaronson, S. Rashed, M. P. Biber, J. A. Hobson, “Brain State and Body Position: A Time-lapse Video Study of Sleep,” Archives of general psychiatry, vol. 39, no. 3, pp. 330-335, 1982.
[22]C. Sahlin, E. Svanborg, H. Stenlundand, K. A. Franklin, “Cheyne-Stokes respiration and supine dependency,” European Respiratory Journal, vol. 25, pp. 829-833, 2005.
[23]R. C. Martin-Du Pan, R. Benoit, L. Girardier, “The role of body position and gravity in the symptoms and treatment of various medical diseases,” Swiss Medicine Weekly, vol. 134, pp. 543-551, 2004.
[24]J. A. Taylor, J. W. Krieger, D. T. Reay, R. L. Davis, R. Harruff, L. K. Cheney., “Prone sleep position and the sudden infant death syndrome in King County, Washington: A case-control study,” The Journal of pediatrics, vol. 128, no. 1, pp. 626-630, 1996.
[25]J. A. Hobson, T. Spagna, R. Malenka, “Ethology of sleep studied with time-lapse photography: postural immobility and sleep-cycle phase in humans,” Science, vol. 201, no. 4363, pp. 1251-1253, 1978.
[26]S. V. Argyropoulos, J. A. Hicks, J. R. Nash, C. J. Bell, A. S. Rich, D. J. Nutt and S. J. Wilson, “Correlation of subjective and objective sleep measurements at different stages of the treatment of depression,” Psychiatry Researchs, vol. 120, no. 2, pp. 179-190, 2009.
[27]S. L. Lin, S. L. Tsai, “The Reliability and Validity of the VSH Sleep Scale Chinese Version,” VGH Nursing, vol. 20, no. 1, pp. 105-106, 2003.
[28]M. A. Grandner, D. F. Kripke, In-Young Yoon, and S. D. Youngstedt, “Criterion Validity of the Pittsburgh Sleep Quality Index: Investigation in a non-clinical sample,” Sleep and Biological Rhythms, vol. 4, pp. 129-136, 2006.
[29]Y. Kishimoto, A. Akahori and K. Oguri, “Estimation of sleeping posture for M-Health by a wearable tri-axis accelerometer,” Proc. of the 3rd IEEE EMBS, pp. 45-48, 2006.
[30]K. V. Laerhoven, M. Borazio, D. Kilian, and B. Schiele, “Sustained Logging and Discrimination of Sleep Postures with Low-Level, Wrist-Worn Sensors,” Proc. Ot the 12th IEEE Intl., Symposium on Wearable Computers (ISWC 2008), pp. 69-77, 2008.
[31]W. T. Ang, P. K. Khosla, and C. N. Riviere, “Nonlinear regression model of a low-g MEMS accelerometer,” IEEE Sensors Journal, vol. 7, no. 1, pp. 81-88, 2007.
[32]A. K. Jain, M. N. Murty, and P. J. Flynn, “Data clustering: A review,” ACM Computing Surveys, vol. 31, no. 3, pp. 264-323, 1999.
[33]江忠綾，陳錫中，白其卉，車先蕙，李明濱，賴仕涵，周碧瑟，“中文版雅典失眠量表信效度研究，”臺灣精神醫學，第23卷，第1期，43-52頁，2009年。
[34]R. E. Gliklich, P. C. Wang, “Validation of the Snore Outcomes Survey for patients with sleep-disordered breathing,” Archives of Otolaryngology-Head & Neck Surgery, vol. 128, pp. 819-824, 2002.
[35]N. H. Chen, H. Y. Li, R. E. Gliklich, C. C. Chu, S. C. Liang, P. C. “Wang, Validation assessment of the Chinese version of the Snore Outcome Survey (SOS),” Qual Life Res, vol. 11, no.6, pp. 601-607,2002.
[36]R. Snyder-Halpern and J. A. Verran, “Instrumentation to Describe Subjective Sleep Characteristics in Healthy Subjects,” Research in Nursing & Health, vol.10, no. 3, pp. 155-163, 1987.
[37]林小玲，蔡欣玲，“「維辛式睡眠量表」中文版之信效度，”榮總護理，第20卷，第1期，105-106頁，2003年。
[38]周文賢，統計學，智勝文化事業有限公司，2007。
[39]Online available: http://140.128.36.80/tfchen/design_fs/c3_anova_post.pdf
[40]B. Hood, D. Bruck and G. Kennedy, “Determinants of Sleep quality in the healthy aged: the role of physical, psychological, circadian and naturalistic light variables,” Age and Ageing, vol. 33, no. 2, pp. 159-165, 2004.