(3.236.231.61) 您好!臺灣時間:2021/05/11 20:40
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

我願授權國圖
: 
twitterline
研究生:何恭宇
研究生(外文):Kung Yu Ho
論文名稱:利用心電圖資料進行睡眠階段辨別之方法研究
論文名稱(外文):A study of ECG-based screening tool for sleep quality assessment
指導教授:林仲志林仲志引用關係
指導教授(外文):C. C. Lin
學位類別:碩士
校院名稱:長庚大學
系所名稱:資訊工程學研究所
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2009
畢業學年度:97
論文頁數:97
中文關鍵詞:睡眠品質心律變異率心電圖類神經網路
外文關鍵詞:sleep qualityHeart Rate Variabilityelectrocardiogramartificial neural network
相關次數:
  • 被引用被引用:0
  • 點閱點閱:682
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:1
本研究嘗試以睡眠中的心電圖(electrocardiogram, ECG)資料為基礎,藉由分析睡眠時期由自律神經平衡的變化對心臟跳動頻率的影響,利用心率變異率(Heart Rate Variability, HRV)於時域部份(5項)以及頻域部份(8項)的各項特徵值,透過倒傳遞類神經網路進行包含(1)甦醒、(2) REM、(3)NREM stage1、(4)NREM stage2、(5)熟睡(NREM stage3與NREM stage4)5項睡眠階段的辨別。期望藉由不同睡眠階段的分析辨別,進而建立(1)總睡眠時間、(2)睡眠潛伏期、(3)睡眠效率、(4)夜裡覺醒次數、(5)快速眼動期與熟睡階段持續時間等5項睡眠品質評估指標。
本研究一共規劃了3個實驗:(1)利用統計的方式驗證本研究所選用的特徵值參數是否具備辨別不同睡眠階段的能力。透過one-way ANOVA檢定證實各項特徵值參數於各睡眠階段皆達到顯著差異(p<0.001)。(2)比較不同網路架構與訓練參數設定的睡眠階段辨別網路其對於各睡眠階段之Youden index,決定較合適的網路訓練參數設定。經實驗測試結果顯示,相對較佳的隱藏層神經元數目設定為30;性能目標(Sum Squared Error, SSE)設定為20。(3)透過特徵參數選取方法篩選出較具睡眠階段辨別能力的輸入參數,減少網路輸入維度以降低網路複雜度。輸入參數維度最終從13維逐步刪減至8維,各睡眠階段的辨別正確率:甦醒階段可達62.83%;REM階段可達59.43%;Stage1階段可達93.33%;Stage2階段可達77.79%;熟睡階段可達81.75%;網路訓練時間為1350.7秒,保留下來的時域特徵值參數為: SDNN、RMSSD、SDSD、NN50、pNN50;頻域特徵值參數為:HF norm、VLF%、5mins total power。
Sleep is traditionally divided into five repeating stages: consciousness, rapid eye movement (REM) sleep, non-rapid eye movement (NREM) sleep stage N1, NREM stage N2, and deep sleep (NREM stage N3 and NREM stage N4). In this study, We established a sleep-stage identification model based on the autonomic heart rate variability data (HRV) acquired by electrocardiogram (ECG). Five different features in HRV duration time and eight features in its frequency were recorded during each stage of sleep and analyzed by the backpropagation artificial neural network to establish a quantitative sleep-quality evaluation standard. It is composed of 1) sleep duration, 2) sleep latency, 3) sleep efficiency, 4) arousal times during sleep, 5) duration of REM and deep sleep.
The model was developed through the following logic: 1) evaluating the effectiveness of selected features in distinguishing stages of sleep by one way analysis of variance (ANOVA) test (statistic significance was set to p<0.001). 2) Fine adjustment of the parameters including learning rate, mean squared error and so on. 3) Diminishing the dimension of the neural network. The results showed that the best architecture of neural network has 8 input nodes and 30 nodes in hidden layers. The sum squared error is set as 20. The distinguishing accuracy of deep sleep is 81.75%. The accuracy of NREM stage1, stage2, REM and wake are 93.33%, 77.79% 59.43% and 62.83% respectively, training time in the artificial neural network is 1350.7 second. Retained features of time domain include SDNN, RMSSD, SDSD, NN50 and pNN50. The HF norm, VLF% and 5mins total power are three retained features in frequency domain.
目錄
誌謝 i
中文摘要 ii
英文摘要 iv
目錄 vi
表目錄 viii
圖目錄 ix
第一章、緒論 - 1 -
1.1研究背景 - 1 -
1.2 研究動機 - 4 -
1.3研究目的 - 6 -
第二章、相關研究 - 9 -
2.1目前利用心電圖進行睡眠階段分析的相關研究 - 9 -
2.2睡眠相關介紹 - 10 -
2.3自律神經系統 - 11 -
2.4心律變異率 - 13 -
2.5 MIT/BIH睡眠資料庫 - 18 -
2.6 類神經網路相關介紹 - 19 -
第三章、研究策略與方法 - 24 -
3.1 睡眠階段辨別 - 25 -
3.2 睡眠階段辨別工具辨別能力評估 - 43 -
3.3 特徵值參數篩選 - 45 -
3.4 睡眠評估指標建立 - 46 -
第四章、實驗方法與結果討論 - 51 -
4.1各項特徵值參數於不同睡眠階段之差異顯著性分析與各項特徵值參數相關性分析實驗 - 51 -
4.2睡眠階段辨別網路參數設定測試實驗 - 57 -
4.3睡眠階段辨別特徵值參數篩選實驗 - 66 -
第五章、結論與未來展望 - 78 -
參考文獻 - 81 -

































表目錄
表一、心律變異率頻譜頻帶定義 - 16 -
表二、頻域分析特徵值(短期記錄分析部分) - 16 -
表三、頻域分析特徵值(長期記錄分析部分) - 17 -
表四、時域分析特徵值 - 17 -
表五、監督式學習類神經網路模型比較 - 22 -
表六、非監督式學習類神經網路模型比較 - 23 -
表七、混淆矩陣 - 44 -
表八、各睡眠階段之分析資料區段筆數 - 52 -
表九、Pearson相關係數涵義 - 54 -
表十、各項特徵值參數之one-way ANOVA分析表 - 55 -
表十一、各項特徵值參數之Pearson相關係數分析表 - 56 -
表十二、各特徵值參數與其具高相關性的特徵值參數 - 56 -
表十三、網路訓練參數設定 - 60 -
表十四、訓練資料筆數與測試資料筆數 - 61 -
表十五、各網路架構與性能目標組合網路的Youden index - 65 -
表十六、各12維輸入參數網路的Youden index - 71 -
表十七、各11維輸入參數網路的Youden index - 72 -
表十八、各10維輸入參數網路的Youden index - 73 -
表十九、各9維輸入參數網路的Youden index - 74 -
表二十、各8維輸入參數網路的Youden index - 74 -
表二十一、各7維輸入參數網路的Youden index - 76 -
表二十二、各維度的最佳網路之辨識正確率 - 77 -














圖目錄
圖一、解波動分析法流程 - 9 -
圖二、睡眠週期圖 - 11 -
圖三、自律神經系統功能圖 - 12 -
圖四、心律變異率與年齡、性別關係圖 - 14 -
圖五、心律變異率頻帶範圍 - 15 -
圖六、(a)單層前饋式類神經網路、(b)多層前饋式類神經網路 - 20 -
圖七、回饋式類神經網路 - 21 -
圖八、監督式學習運作流程 - 21 -
圖九、非監督式學習運作流程 - 23 -
圖十、系統流程圖 - 24 -
圖十一、睡眠階段辨識功能架構圖 - 25 -
圖十二、分析資料區段切割方式示意圖 - 27 -
圖十三、“So and Chan”R波偵測流程圖 - 30 -
圖十四、正切雙彎曲轉移函數 - 34 -
圖十五、對數雙彎曲轉移函數 - 35 -
圖十六、線性轉移函數 - 35 -
圖十七、Sequential Backward Selection流程圖 - 46 -
圖十八、各項特徵值參數於不同睡眠階段之差異顯著性分析與各項特徵值參數相關性分析實驗流程圖 - 53 -
圖十九、睡眠階段辨別網路參數設定測試實驗步驟流程圖 - 62 -
圖二十、睡眠階段辨別特徵值參數篩選實驗步驟流程圖 - 68 -
參考文獻
[1] 鄭月雅,由社會發展趨勢探討國人睡眠品質,行政院主計處中部辦公室,http://www.dgbas.gov.tw/public/Data/4122719141571.pdf。
[2] Dodel R, Peter H, Spottke A, Noelker C, Althaus A, Siebert U, Walbert T, Kesper K, Becker HF, Mayer G, “Health-related quality of life in patients with narcolepsy”, Sleep Med, 2007.
[3] Vourdas A, Shneerson JM, Gregory CA, Smith IE, King MA, Morrish E, “Narcolepsy and psychopathology: Is there an association? ”, Sleep Med, Vol.3(4), pp.353-60, 2002.
[4] Bixler EO, Vgontzas AN, Lin HM, Calhoun SL, Vela-Bueno A, Kales A, “Excessive Daytime Sleepiness in a General Population Sample: The Role of Sleep Apnea, Age, Obesity, Diabetes and Depression”, J Clin Endocrinol Metab, Vol.90(8), pp.4510-4515, 2005.
[5] Bittencourt LR, Silva RS, Santos RF, Pires ML, Mello MT, “Excessive daytime sleepiness”, Rev Bras Psiquiatr, Vol.27(1), pp.16-21, 2005.
[6] 行政院主計處,94年度社會發展趨勢調查-健康安全, http://www.dgbas.gov.tw/mp.asp?mp=1。
[7] 陳明輝、楊建銘,「眾人皆睡,我獨醒?」--淺談老年人的失眠問題與因應,中興醫院睡眠中心,http://cs.csh.org.tw/into/into02/%BA%CE%AFv%A4%A4%A4%DF/%A4%DF%B2z%BF%C5%C5%B2%AB%C7/%A1u%B2%B3%A4H%AC%D2%BA%CE%A1A%A7%DA%BFW%BF%F4%A1H%A1v--%B2L%BD%CD%A6%D1%A6~%A4H%AA%BA%A5%A2%AFv%B0%DD%C3D%BBP%A6%5D%C0%B3%A1%5D%AE%FC%B3%F8%A5%5B%A4u%A1%5E.doc。
[8] 鄭智銘,睡眠障礙與睡眠評估,元智大學老人福祉科技研發中心,
http://designer.mech.yzu.edu.tw/article/articles/technical/file/(2002-07-20)%20%BA%CE%AFv%BB%D9%C3%AA%BBP%BA%CE%AFv%B5%FB%A6%F4.pdf,2002。
[9] 鄭智銘,睡眠監測相關研究簡介,元智大學老人福祉科技研發中心,2002。
[10] R.Lerner, “Sleep loss in the aged: implications for nursing practice”, journal of Gerontological Nursing, vol. 8, no. 6, pp. 323-326, 1982
[11] D.Cohen, C. Eisdorfer, P. Prinz, A. Breen, M. Davis, and A. Gadsby, “Sleep disturbances in the institutionalized aged”, Journal of the American Geriatrics Society, vol. 31, no. 2, pp. 79-82, 1993.
[12] He´le`ne Otzenberger, Claude Gronfier, Chantal Simon, Anne Charloux, Jean Ehrhart, Franc¸ Ois Piquard, Gabrielle Brandenberger, “Dynamic heart rate variability: a tool for exploring sympathovagal balance continuously during sleep in men”, American Physiological Society, pp.946-950, 1998.
[13] Gabrielle Brandenberger, Antoine U. Viola, Jean Ehrhart, Anne Charloux, Bernard Geny, Franc¸ Ois Piquard, Chantal Simon, “Age-related changes in cardiac autonomic control during sleep”, J. Sleep Res., 12, pp. 173–180, 2003.
[14] Tetsuya Miyashita, Kenichi Ogawa, Hideki Itoh, Yukiko Arai, Minako Ashidagawa, Munehito Uchiyama, Yasuhiro Koide, Tomio Andoh, Yoshitsugu Yamada, “Spectral analyses of electroencephalography and heart rate variability during sleep in normal subjects”, Autonomic Neuroscience: Basic and Clinical, 103, pp. 114– 120, 2003.
[15] MINA AKO, TOKUHIRO KAWARA, SUNAO UCHIDA, SHINICHI MIYAZAKI, KYOKO NISHIHARA, JUNKO MUKAI, KENZO HIRAO, JUNYA AKO, YOSHIRO OKUBO, “Correlation between electroencephalography and heart rate variability during sleep”, Psychiatry and Clinical Neurosciences, 57, pp. 59–65, 2003.
[16] Terry B. J. Kuo, Cheryl C. Yang, “Scatterplot Analysis of EEG Slow-Wave Magnitude and Heart Rate Variability: An Integrative Exploration of Cerebral Cortical and Autonomic Functions”, SLEEP, Vol. 27, No. 4, pp.648-656, 2004.
[17] P. BUŠEK, J. VAŇKOVÁ, J. OPAVSKÝ1, J. SALINGER1, S. NEVŠÍMALOVÁ, ”Spectral Analysis of Heart Rate Variability in Sleep”, Physiol. Res. 54, pp369-376, 2005.
[18] Thomas Penzel, Member, IEEE, Jan W. Kantelhardt, Ludger Grote, Jörg-Hermann Peter, and Armin Bunde,” 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, OCTOBER 2003.
[19] F. Jurysta, P. van de Borne, P.-F. Migeotte, M. Dumont, J.-P. Lanquart, J.-P. Degaute, P. Linkowski, “A study of the dynamic interactions between sleep EEG and heart rate variability in healthy young men”, Clinical Neurophysiology 114 , p.p.2146-2155, 2003.
[20] Raffaele Ferri, Liborio Parrino, Arianna Smerieri, Mario G. Terzano, Maurizio Elia, Sebastiano A. Musumeci, Salvatore Pettinato, “Cyclic alternating pattern and spectral analysis of heart rate variability during normal sleep”, European Sleep Research Society, pp.13-18, 2000.
[21] F. Jurysta, J.-P. Lanquart, P. van de Borne, P.-F. Migeotte, M. Dumont, J.-P. Degaute, P. Linkowski1, “The link between cardiac autonomic activity and sleep delta power is altered in men with sleep apnea-hypopnea syndrome”, Am J Physiol Regul Integr Comp Physiol, 291, pp.1165–1171, 2006.
[22] M.H. Bonnet, D.L. Arandb, “Heart rate variability: sleep stage, time of night, and arousal influences”, Electroencephalography and clinical Neurophysiology, 102, p.p.390-396, 1997.
[23] Gih Sung Chung, Byoung Hoon Choi, Ko Keun Kim, Yong Gyu Lim, Jin Wook Choi, DO-Un Jeong, Kwang Suk Park, “Unconstrained Heart Rate Estimation System in Bed”, International Conference on Convergence Information Technology, pp.2363-2367, 2007.
[24] Ya-Ti Peng, Student Member, IEEE, Ching-Yung Lin, Senior Member, IEEE, Ming-Ting Sun, Fellow, IEEE, and Carol A. Landis, ” Multimodality Sensor System for Long-Term Sleep Quality Monitoring”, IEEE Transactions on Biomedical Circuits and Systems, VOL. 1, NO. 3, pp.217-227, SEPTEMBER 2007.
[25] Terry B. J. Kuo, Fu-Zen Shaw, Ching J. Lai, Chery C. H. Yang, “Asymmetry in Sympathetic and Vagal Activities During Sleep-Wake Transitions”, SLEEP, Vol. 31, No. 3, pp.311-320, 2008
[26] 台灣科學地-心率大師-ANSWatch-什麼是HRV, http://www.taiwanscientific.com.tw/WhatIsHRV.html。
[27] 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”, European Heart Journal 17, pp. 354-381, 1996.
[28] MT/BIH physioBank, http://www.physionet.org .
[29] A. Rechtschaffen and A. Kales, A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stage of Human Subject, Washington, DC: US Government Printing Office, 1968.
[30] 張斐章、張麗秋,《類神經網路》,初版,台北,東華書局,民國九十六年八月。
[31] H.H. So, K.L. Chan. “Development of QRS Detection Method for Real-Time Ambulatory Cardiac Monitor”. In Proceedings of the 19th Annual International Conference of the IEEE in Engineering in Medicine and Biology Society, Vol. 1, pp. 289-292, 1997.
[32] K. F. Tan, K. L. Chan, and K. Choi, “Detection of the QRS Complex, P Wave and T Wave in Electrocardiogram”, IEE Conference Publication , no. 476, pp. 41-47. 2000.
[33] 羅華強,《類神經網路-MATLAB的應用》,二版,台北,高立圖書,民國九十七年元月。
[34] MacKay, D.J.C., “A Practical Bayesian Framework for Backpropagation Network”, Computation, V.4, No.3, pp.448-472, 1992.
[35] Enrique F. Schisterman, Neil J. Perkins, Aiyi Liu, Howard Bondell, “Optimal Cut-point and Its Corresponding Youden Index to Discriminate Individuals Using Pooled Blood Samples”, Epidemiology, Vol 16, No. 1, pp. 73-81, January 2005
[36] 吳宏達,統計學方法在醫學診斷上之應用(1),中國醫藥大學生物統計中心, http://www2.cmu.edu.tw/~biostat/online/teaching_corner_001.pdf。
[37] 楊敦翔,〈以類神經網路與特徵選取技巧處理空氣能見度預測問題之研究〉,國立中山大學機械與機電工程研究所,碩士論文,民國92年7月。
[38] 林震岩,《多變量分析-SPSS的操作與應用》,台北,智勝文化,民國九十六年元月。
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊
 
系統版面圖檔 系統版面圖檔