進階搜尋


 
系統識別號 U0026-0812200914323673
論文名稱(中文) 運用資料探勘技術由腦波訊號建立喚醒偵測模型
論文名稱(英文) Construction of Arousal Detection Models from EEG Signals by Using Data Mining Techniques
校院名稱 成功大學
系所名稱(中) 醫學資訊研究所
系所名稱(英) Institute of Medical Informatics
學年度 96
學期 2
出版年 97
研究生(中文) 趙茂偉
研究生(英文) Mao-wei Chao
電子信箱 q5695108@mail.ncku.edu.tw
學號 q5695108
學位類別 碩士
語文別 中文
論文頁數 71頁
口試委員 指導教授-曾新穆
口試委員-蕭富仁
口試委員-孫永年
口試委員-梁勝富
中文關鍵字 腦波圖  睡眠呼吸中止  基因演算法  分類器  喚醒  傅立葉轉換 
英文關鍵字 Sleep Apnea  Electroencephalogram  Fast Fourier Transform  Classification  Genetic Algorithms  Arousal 
學科別分類
中文摘要 近年來,睡眠呼吸中止症(Sleep Apnea Syndrome)是常見的睡眠疾病,其症狀為患者在睡眠中會出現頻繁的呼吸中止,進而引起大腦的缺氧,因而影響健康,嚴重的病患亦可能導致死亡。根據文獻,睡眠呼吸中止後腦波會發生頻帶的轉移現象,此現象稱為喚醒(Arousal),可當作睡眠呼吸中止的腦波特徵。故在本論文中,我們首先分析分類器與喚醒之間的關係。然後,我們提出以基因演算法為基礎之方法進行睡眠腦波喚醒的混合式分類器偵測模型建置。所提的方法中,首先使用快速傅立葉轉換(Fast Fourier Transform)進行不同頻帶的腦波訊號擷取。接著,從所得的腦波訊號再進行特徵擷取後,經由所提的方法建置睡眠腦波喚醒偵測的混合式分類器模型。因此,在所提的方法中每條染色體表示ㄧ個可能的混合式分類器模型。而在評估函數(fitness function)部分,我們使用準確度(Precision)、涵蓋度(Recall)和F-measure三種評估函數進行染色體的適合度(fitness)的計算。最後,最好的染色體即為我們所得到的最佳混合式分類器模型。實驗部分,我們使用十六位患有睡眠呼吸中止症的腦波進行分析。首先,使用單一分類器和單一分類器整合效能提升演算法來分析喚醒與分類模型之間的關係。最後,針對所提的方法進行實驗的驗證,實驗結果證實由所提的方法不但可以得出適合的睡眠腦波喚醒偵測模型,而且在模型效能上亦有顯著的改善。
英文摘要 Sleep apnea is a common sleep disorder, the symptom is that patients may show respirating pause frequently during sleep such that cause brain injury due to lack of oxygen and result in poor health. A considerable portion of cases may even cause death. According to literatures, when sleep apnea syndrome happens, electroencephalogram (EEG) may show shifting phenomena on the frequency band, particularly inducing “Arousal”. Arousal can thus be considered as the features of sleep apnea on EEG. In this thesis, we first analyze the relationship between arousal and classification model. We then propose a GA(Genetic Algorithms)-based approach for building a hybrid classification model for arousal detection. In the proposed approach, the Fast Fourier transform (FFT) is first used to convert the EEG signal into frequency domain. Features are then extracted from these transformed data and used as classification attributes in the proposed approach. Thus, each chromosome represents a possible hybrid classification model for arousal detection. Three fitness functions, including Precision, Recall and F-measure, are used to evaluate the fitness of a chromosome. After the evolution processing, the best chromosome is then output as the final hybrid classification model. In the experiments, EEG signals of the sixteen patients that have sleep apnea syndrome are used for analyzing. First, the experiments were made to show the relationship among single classifier and single classifier with improved approach and arousal. The experiments were then made to verify the proposed approach and the results show that the proposed approach can not only derived appropriate hybrid classification model, but also has significant improvement on the accuracy of the classification model.
論文目次 中文摘要 I
英文摘要 II
目錄 III
表目錄 V
圖目錄 VI
第一章 導論 1
1.1 研究背景 1
1.2 研究動機 1
1.3 問題描述 2
1.4 研究方法 3
1.5 論文架構 4
第二章 文獻探討 5
2.1 ASDA喚醒的判定規則 5
2.2 腦波圖(Electroencephalogram)的應用領域 5
2.3 睡眠呼吸中止症 6
2.4 口鼻腔氣流、胸部腹部呼吸運動、血氧飽和度、心電圖 7
2.5 睡眠腦波 8
2.6 分類器 9
2.6.1 分類樹探勘法(Classification Mining on Decision Tree) 9
2.6.1.1 分類樹探勘法之定義 10
2.6.1.2 分類樹探勘法之目的 11
2.6.1.3 分類樹探勘演算法說明 12
2.6.1.4 WEKA J48演算法說明 13
2.7 單一分類器整合效能提升演算法 13
2.7.1 Bagging 13
2.7.2 AdaBoost 14
2.7.3 Random Subspace 18
2.8 基因演算法 20
2.8.1 基因演算法簡述 20
2.8.2 基因演算法的設計 21
第三章 資料探勘方法建置偵測喚醒之模型 24
3.1 方法架構描述 24
3.2 使用基因演算法實作混合式分類器(Hybrid Classifiers) 25
3.2.1 訊號處理與特徵擷取 25
3.2.2 染色體編碼 28
3.2.3 適應函數 30
3.2.4 基因運算子 32
3.2.5 選擇 33
3.3 所提之演算法 34
3.4 混合式分類器範例 36
第四章 實驗結果 39
4.1 實驗資料 39
4.2 實驗設計 42
4.2.1 單一分類器 43
4.2.2 單一分類器整合效能提升演算法 43
4.2.3 所提的方法 44
4.3 實驗分析 45
4.3.1 單一分類器 45
4.3.2 單一分類器整合效能提升演算法 49
4.3.3 使用基因演算法建構一個混合式分類器 52
4.4 實驗總結 59
第五章 系統建構 61
5.1 系統介面和功能說明 61
第六章 結論與未來研究方向 67
6.1 結論 67
6.2 應用與研究方向 68
參考文獻 69
參考文獻 [1] C.W. Anderson, “Determining Mental State from EEG Signals Using Neural Networks,” Scientific programming, pp. 171-183, 1995.
[2] Paulo Aguiar, “EEGSolver-Brain Activity and Genetic Algorithms,” ACM Press, pp. 80-84, 2000.
[3] L. Breiman, “Bagging Predictiors,” Machine Learning, pp. 123-140, 1996.
[4] J. C. Burges, “A Tutorial on Support Vector Machines for Pattern Recognition,” Data Mining and Knowledge Discovery, pp. 121-167, 1998.
[5] Joseph J. Carr, “Introduction to Biomedical Equipment Technology 4th ed.,” Prentice Hall, pp. 369-392, 2001.
[6] P. De Chazal, C. Heneghan, E. Sheridan, R. Reilly and P. Nolan, M. O'Malley, “Automatic Classification of Sleep Apnea Epochs Using the Electrocardiogram,” Computers in Cardiology, pp. 745–748, 2000.
[7] S. P. Cho, “Detection of Aousal in Patients with Respiratory Sleep Disorders Using a Single Channel EEG,” Medicine and Biology Annual Conference, pp. 2733-2735, 2005.
[8] M. J. Drinnan, “Automated Recognition of EEG Changes Accompanying Arousal in Respiratory Sleep Disorders,” Sleep, pp. 296-303, 1996.
[9] P. Domingos, “On the Optimality of the Simple Bayesian Classifier Under Zero-one Loss,” Machine Learning, pp. 103.130, 1997.
[10] E. Frank, M. A. Hall, Geoffrey Holmes, Richard Kirkby, Bernhard Pfahringer and Ian H, “Weka – a Machine Learning Workbench for Data Mining,” The Data Mining and Knowledge Discovery Handbook, pp. 1305-1314, 2005.
[11] Y. Freund and R. E. Schapire, “Experiments with A New Boosting Algorithm,” Machine Learning, pp. 148-156, 1996.
[12] T. K. Ho, “The Random Subspace Method for Constructing Decision Forests,” IEEE Transactions on Pattern Analysis and Machine Intelligence, pp. 832-844, 1998.
[13] H. H. Jasper, “The Ten-twenty Electrode System of the International Federation,” Electroencephalography and Clinical Neurophysiology, pp. 371-375, 1958.
[14] M. Kermit, A. J. Eide and T. Lindblad, K. Waldemark, “Treatment of Obstructive Sleep Apnea Syndrome by Monitoring Patients Airflow Signals,” Pattern Recognition Letters, pp.277-281, 2000.
[15] J. E. Mietus, “Detection of obstructive sleep apnea from cardiac interbeat interval time series,” Computers in Cardioligy, pp. 753-756, 2000.
[16] Z. Michalewicz, “Genetic Algorithms + Data Structures = Evolution Programs,” Springer-Verlag, 1994.
[17] J. R. Quinlan, “Induction of decision tree,” Machine Learning, pp.81-106, 1986.
[18] J. R. Quinlan, “C4.5: Programs for Machine Learning, Morgan Kaufman,” 1993.
[19] Allan Rechtschaffen, “A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects,” CA: UCLA BIS/BRI, 1968.
[20] F. Sartoretto, “Automatic Detection of Epileptiform Activity by Single-level Wavelet Analysis,” Clinical Neurophysiology, pp. 239-247, 1999.
[21] M. Varanunu, A. Mortara , M. Raciti, M.T. La Rovere, R. Maestri, G.D. Pinna, M. Bertinelli and M. Emdin, “Apnea Detection from 24-Hour Recordings of Respiration in Chronic Heart Failure Patients,” Computer in Cardiology, pp.489-492, 2000.
[22] P. Varady, S. Bongar and Z. Benyo, “Detection of Airway Obstruction and Sleep Apnea by Analyzing the Phase Relation of Respiration Movement Signals,” IEEE Instrumentation and Measurement Technology Conference, pp.185-190, 2001.
[23] P. Varady, “On-line Detection of Sleep Apnea During Critical Care Monitoring,” Proceedings of the 22nd Annual EMBS International Conference, pp. 1299-1301, 2000.
[24] V. Vapnik, “The Nature of Statical Learning Theory,” Springer Verlag, 1995.
[25] J. Willis, “The Topography of Muscle Activity in Quantitative EEG,” Clinical Electroencephalography, pp. 123-126, 1993.
[26] John G. Webster, “Medical Instrumentation Application and Design 3rd ed.,” Wiley, pp. 156- 176, 1998.
[27] M. Zamora, “The Study of Micro-Arousals Using Neural,” IEE Conference Publication, pp. 625-630, 1999.
[28] ASDA Report, “EEG Arousal: Scoring Rules and Examples,” Sleep, pp.173-184, 1992.
[29] Colorado Health Site, http://www.coloradohealthsite.org/
[30] Sleep Study, http://www.sleepstudy.org/
[31] 許慶煌,”顳葉痲癇之電腦輔助腦波圖分析,成功大學工程研究所碩士論文”,民國八十七年。
[32] 楊杰,”聽覺誘發電位之電腦化分析系統,中原大學醫學工程研究所碩士論文”,民國八十五年。
論文全文使用權限
  • 同意授權校內瀏覽/列印電子全文服務,於2009-08-25起公開。
  • 同意授權校外瀏覽/列印電子全文服務,於2010-08-25起公開。


  • 如您有疑問,請聯絡圖書館
    聯絡電話:(06)2757575#65773
    聯絡E-mail:etds@email.ncku.edu.tw