博碩士論文 etd-0803119-103054 詳細資訊


[回到前頁查詢結果 | 重新搜尋]

姓名 吳登翊(Deng-Yi Wu) 電子郵件信箱 E-mail 資料不公開
畢業系所 電機工程學系研究所(Department of Electrical Engineering)
畢業學位 碩士(Master) 畢業時期 107學年第2學期
論文名稱(中) 機器學習應用於心電圖心跳模式之分類研究
論文名稱(英) Applying Machine Learning on Classification of Electrocardiogram Heartbeat Patterns
檔案
  • etd-0803119-103054.pdf
  • 本電子全文僅授權使用者為學術研究之目的,進行個人非營利性質之檢索、閱讀、列印。
    請遵守中華民國著作權法之相關規定,切勿任意重製、散佈、改作、轉貼、播送,以免觸法。
    論文使用權限

    紙本論文:5 年後公開 (2024-09-03 公開)

    電子論文:使用者自訂權限:校內 5 年後、校外 5 年後公開

    論文語文/頁數 中文/50
    統計 本論文已被瀏覽 5644 次,被下載 0 次
    摘要(中) 心電圖(ECG)記錄了患者心臟的生理信號。醫生可以透過觀察並以心電圖來判斷給出相應的症狀。對於大量的心電圖數據,單純只依靠醫生的診斷是非常沒有效率的。此外,每位醫生的主觀判斷也可能不同,在疲勞的狀態下也可能會影響準確性判斷。機器學習技術的應用可以有效減少醫生的工作和誤診。在本論文中,我們提出了一種建構徑向基底函數網路(RBF)的方法用於分類心電圖心跳模式。該方法由五部分組成。首先,我們將ECG從圖片轉換為數位資料。第二,利用斜率來偵測完整的PQRST波形,並移除ECG中的非心跳部分。第三,使用插值方法將不同長度的心跳轉換成相同長度,解決模型需要相同輸入維度的問題。第四,使用分群技術對心跳模式進行分群,並將由此產生的群聚作為後續模型的隱藏層參數使用。最後,我們使用最小二乘法來找到隱藏層與輸出層間關聯的最佳權重值。我們提出的方法有下列的優點:(1)從大量的心電圖中自動偵測完整心跳位置,可省去許多人工處理的時間;(2)利用插值法可以將不同長度轉換為相同長度,並且不破壞波形的樣子,藉此解決訓練模型時需要相同長度的問題;(3)藉由分群法能夠自動得出相關性較大的各群聚,不須使用者自定義。由我們的實驗結果可知,我們提出的方法能夠有不錯的效果,準確度有高達95%,其餘指標也相比其他研究還來的好。
    摘要(英) The electrocardiogram (ECG) records the physiological signals of the patient. The physician can judge and give the corresponding symptoms by observing the electrocardiogram. For a large number of ECG data, it is very inefficient to rely solely on physician diagnosis. In addition, the subjective judgments of doctors may also be different, and fatigue may also affect the accuracy of judgment. The application of machine learning technology can effectively reduce the work and misdiagnosis of doctors. In this paper, we propose a method to construct radial basis function (RBF) networks for classifying ECG heartbeat patterns. The method consists of five parts. First, we convert the ECG from a picture to a digital file. Second, the non-heartbeat components in the ECG are removed using the slope to extract the complete PQRST waveform. Third, the interpolation method is used to convert the heartbeats of different lengths into the same length, solving the problem that the model needs the same input dimension. Fourth, a clustering technique is used to cluster the heartbeat patterns, and the resulting clusters form the radial basis functions of the hidden layer. Finally, we use the least square method is used to find the optimal values of the weights associated with the output layer. The proposed method has the following advantages: (1) automatically detecting the complete heartbeat position from a large number of electrocardiograms, which can save a lot of manual processing time; (2) The interpolation method can be used to convert different lengths into the same length without destroying the waveform, thereby solving the problem of requiring the same length when training the model; (3) By clustering method, each cluster with greater correlation can be automatically obtained without user customization. From our experimental results, we can see that the proposed method can have a good result, the accuracy is as high as 95%, and the other indicators are better than other studies.
    關鍵字(中)
  • 心電圖
  • 分群
  • 分類
  • 機器學習
  • 徑向基底函數
  • 神經網路
  • 插值法
  • 關鍵字(英)
  • Neural network
  • Classification
  • Clustering
  • Machine learning
  • Radial basis function
  • Interpolation
  • Electrocardiogram
  • 論文目次 論文審定書 i
    誌謝 ii
    摘要 iii
    圖目錄 vii
    表目錄 viii
    第一章 導論 1
    1.1. 研究背景與目的 1
    1.2. 心電圖概述 2
    1.3. 心電圖數據資料 4
    1.4. 論文架構 8
    第二章 文獻探討 9
    2.1. 自建構分群(Self-Constructing Clustering, SCC) 9
    第三章 研究方法 11
    3.1. 心電圖圖檔數位化 12
    3.2. 波形偵測 14
    3.3. 以內插法將不同長度心跳轉換為相同長度 17
    3.4. 透過迭代自建構分群SCC-I進行分群 20
    3.5. 最小二乘法求RBF模型權重 22
    3.6. 預測病患的病徵 23
    第四章 實驗結果 24
    4.1. 測量指標 24
    4.2. MIT-BIH心律失常數據庫實驗 25
    4.3. 高醫提供的心電圖 28
    4.3.1. 心電圖數位化 28
    4.3.2. 波形偵測結果展示 29
    4.3.3. 內插法結果展示 30
    4.3.4. 使用高醫心電圖數據將RBF模型與其他分類器進行比較 31
    第五章 結論與未來展望 34
    5.1. 結論 34
    5.2. 未來研究方向 34
    參考文獻 35
    參考文獻 [1] Y. N. Narain and P. Gupta, “ECG to individual identification,” 2008 IEEE Second International Conference on Biometrics: Theory, Applications and Systems, IEEE, 2008.
    [2] F. Sufi, I. Khalil and I. Habib, “Polynomial distance measurement for ECG based biometric authentication," Security and Communication Networks, vol. 3, no. 4, pp. 303-319, 2010.
    [3] C. Li, C. Zheng and C. Tai, “Detection of ECG characteristic points using wavelet transforms,” IEEE Transactions on biomedical Engineering, vol. 42, no. 1, pp. 21-28, 1995.
    [4] F. Gritzali, G. Frangakis and G. Papakonstantinou, “Detection of the P and T waves in an ECG,” Computers and Biomedical Research, vol. 22, no. 1, pp. 83-91, 1989.
    [5] Q. Xue, Y. H. Hu and W. J. Tompkins, “Neural-network-based adaptive matched filtering for QRS detection,” IEEE Transactions on Biomedical Engineering, vol. 39, no. 4, pp. 317-329, 1992.
    [6] P. S. Hamilton and W. J. Tompkins, “Quantitative investigation of QRS detection rules using the MIT/BIH arrhythmia database,” IEEE transactions on biomedical engineering, vol. BME-33, no. 12, pp.1157-1165, 1986.
    [7] E. Pietka, “Feature extraction in computerized approach to the ECG analysis,” Pattern Recognition, vol. 24, no. 2, pp. 139-146, 1991.
    [8] M. Kropf, D. Hayn and G. Schreier, “ECG classification based on time and frequency domain features using random forests,” 2017 Computing in Cardiology, IEEE, 2017.
    [9] N. Emanet, “ECG beat classification by using discrete wavelet transform and Random Forest algorithm,” 2009 Fifth International Conference on Soft Computing, Computing with Words and Perceptions in System Analysis, Decision and Control, IEEE, 2009.
    [10] J. Park, S. Lee and K. Kang, “Arrhythmia detection using amplitude difference features based on random forest,” 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, IEEE, 2015.
    [11] A. Walinjkar and J. Woods, “ECG classification and prognostic approach towards personalized healthcare,” 2017 International Conference On Social Media, Wearable And Web Analytics, IEEE, 2017.
    [12] P. Shimpi, S. Shah, M. Shroff and A. Godbole, “A machine learning approach for the classification of cardiac arrhythmia,” 2017 International Conference on Computing Methodologies and Communication, IEEE, 2017.
    [13] S. Faziludeen and P. Sankaran, “ECG beat classification using evidential K-nearest neighbours,” Procedia Computer Science, vol. 89, pp. 499-505, 2016.
    [14] M. Soliński, A. Perka, J. Rosiński, M. Łepek and J. Rymko, “Classification of atrial fibrillation in short-term ECG recordings using a machine learning approach and hybrid QRS detection,” 2017 Computing in Cardiology, IEEE, 2017.
    [15] H. Lassoued and R. Ketata, “ECG multi-class classification using neural network as machine learning model,” 2018 International Conference on Advanced Systems and Electric Technologies, IEEE, 2018.
    [16] A. Gavhane, G. Kokkula, I. Pandya and K. Devadkar, “Prediction of Heart Disease Using Machine Learning,” 2018 Second International Conference on Electronics, Communication and Aerospace Technology, IEEE, 2018.
    [17] P. Shimpi, S. Shah, M. Shroff and A. Godbole, “A machine learning approach for the classification of cardiac arrhythmia,” 2017 International Conference on Computing Methodologies and Communication, IEEE, 2017.
    [18] C. Liu, Q. Li, P. B. Suresh, A. Vest and G. D. Clifford, “Multi-source features and support vector machine for heart rhythm classification,” 2017 Computing in Cardiology, IEEE, 2017.
    [19] M. S. Refahi, J. A. Nasiri and S. M. Ahadi, “ECG Arrhythmia Classification using Least Squares Twin Support Vector Machines,” Electrical Engineering (ICEE), Iranian Conference on, IEEE, 2018.
    [20] U. R. Acharya, H. Fujita, O. S. Lih, Y. Hagiwara, J. H. Tan and M. Adam, “Automated detection of arrhythmias using different intervals of tachycardia ECG segments with convolutional neural network, Information sciences”, vol. 405, pp. 81-90, 2017.
    [21] S. Kiranyaz, T. Ince and M. Gabbouj, “Real-time patient-specific ECG classification by 1-D convolutional neural networks,” IEEE Transactions on Biomedical Engineering, vol. 63, no. 3, pp. 664-675, 2015.
    [22] M. Limam and F. Precioso, “ation detection and ECG classification based on convolutional recurrent neural network,” 2017 Computing in Cardiology, IEEE, 2017.
    [23] S. Kiranyaz, T. Ince, R. Hamila and M. Gabbouj, “Convolutional neural networks for patient-specific ECG classification,” 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, IEEE, 2015.
    [24] T. Kao, L. J. Hwang, Y. H. Lin, T. H. Lin and C. H. Hsiao, “Computer analysis of the electrocardiograms from ECG paper recordings,” 2001 Conference Proceedings of the 23rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, vol. 4, IEEE, 2001.
    [25] P. R. K. Shrivastava, S. Panbude and G. Narayanan, “Digitization of ECG paper records using MATLAB,” International Journal of Innovative and Exploring Engineering, vol. 4, no. 6, 2014.
    [26] G. A. Virgin and V. V. Baskar, “Conversion of ECG Graph into Digital Format,” International Journal of Pure and Applied Mathematics, vol. 118, no. 17, pp. 469-484, 2018.
    [27] M. S. Islam, N. Alajlan and S. Malek, “Resampling of ECG signal for improved morphology alignment,” Electronics letters, vol. 48, no. 8, pp. 427-429, 2012.
    [28] A. W. V. Hof, A. Liem, M. J. D. Boer, F. Zijistra, Z. M. I. S. Group, “Clinical value of 12-lead electrocardiogram after successful reperfusion therapy for acute myocardial infarction,” The Lancet, vol. 350, no. 9078, pp. 615-619, 1997.
    [29] M. K. Das, H. Suradi, W. Maskoun, M. A. Michael, C. Shen, J. Peng, G. Dandamudi and J. Mahenthiran, “Fragmented wide QRS on a 12-lead ECG: a sign of myocardial scar and poor prognosis,” Circulation: Arrhythmia and Electrophysiology, vol. 1, no. 4, pp. 258-268, 2008.
    [30] H. Nagendra, S. Mukherjee and V. Kumar, “Application of wavelet techniques in ECG signal processing: an overview,” Int J Eng Sci Technol, vol. 3, no. 10, pp. 7432-7443, 2011.
    [31] A. ECAR, “Recommended Practice for Testing and Reporting Performance Results of Ventricular Arrhythmia Detection Algorithms,” Association for the Advancement of Medical Instrumentation, 1987.
    [32] V. M. Guerra, J. Novo, J. Rouco, M. G. Penedo and M. Ortega, “Heartbeat classification fusing temporal and morphological information of ECGs via ensemble of classifiers,” Biomedical Signal Processing and Control, vol. 47, pp. 41-48, 2019.
    [33] J. Pan and W. J. Tompkins, “A Real-Time QRS Detection Algorithm,” IEEE Trans. Biomed. Eng, vol. 32, no. 3, pp. 230-236, 1985.
    [34] Y. C. Yeh and W. J. Wang, “QRS complexes detection for ECG signal: The Difference Operation Method,” peration Method." Computer methods and programs in biomedicine, vol. 91, no. 3, pp. 245-254, 2008.
    [35] Y. Wang, C. J. Deepu and Y. Lian, “A computationally efficient QRS detection algorithm for wearable ECG sensors,” 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, IEEE, 2011.
    [36] C. Xiaomeng, “A new real-time ecg r-wave detection algorithm,” Proceedings of 2011 6th International Forum on Strategic Technology, vol. 2, pp. 1252-1255, IEEE, 2011.
    [37] F. Rezazadeh, H. Seno, “A New Heart Arrhythmia’s Detection Algorithm,” Journal of Knowledge Management, Economics and Information Technology, vol. 3, no. 6, pp. 1-4, 2013.
    [38] C. A. Hall and W. W. Meyer, “Optimal error bounds for cubic spline interpolation,” Journal of Approximation Theory, vol. 16, no. 2, pp. 105-122, 1976.
    [39] E. Meijering, “A chronology of interpolation: from ancient astronomy to modern signal and image processing,” Proceedings of the IEEE, vol. 90, no. 3, pp. 319-342, 2002.
    [40] K. W. Stewart, F. Thomas, C. Pretty, J. G. Chase and G. M. Shaw, “How should we interpret retrospective blood glucose measurements? Sampling and Interpolation,” IFAC-PapersOnLine, vol. 50, no. 1, pp. 874-879, 2017.
    [41] J. Y. Jiang, R. J. Liou and S. J. Lee, “A fuzzy self-constructing feature clustering algorithm for text classification,” IEEE transactions on knowledge and data engineering, vol. 23, no. 3, pp. 335-349, 2010.
    [42] D. W. Marquardt, “An algorithm for least-squares estimation of nonlinear parameters,” Journal of the society for Industrial and Applied Mathematics, vol. 11, no. 2, pp. 431-441, 1963.
    [43] H. B. Demuth, M. H. Beale, O. D. Jess and M. T. Hagan, “Neural network design,” 2014.
    [44] P. D. Chazal and M. O’Dwyer and R. B. Reilly, “Automatic classification of heartbeats using ECG morphology and heartbeat interval features,” IEEE transactions on biomedical engineering, vol. 51, no. 7, pp. 1196-1206, 2004.
    [45] T. Mar, S. Zaunseder, J. P. Martínez, M. Llamedo and R. Poll, “Optimization of ECG classification by means of feature selection,” IEEE transactions on Biomedical Engineering, vol. 58, no. 8, pp. 2168-2177, 2011.
    [46] M. L. Soria and J. P. Martinez, “An ECG classification model based on multilead wavelet transform features,” 2007 Computers in Cardiology, IEEE, 2007.
    [47] J. Cohen, “A coefficient of agreement for nominal scales,” Educational and psychological measurement, vol. 20, no. 1, pp. 37-46, 1960.
    [48] M. Fatourechi, R. K. Ward, S. G. Mason, J. Huggins, A. Schlögl and G. E. Birch, “Comparison of evaluation metrics in classification applications with imbalanced datasets,” 2008 Seventh International Conference on Machine Learning and Applications, IEEE, 2008.
    [49] Z. Zhang, J. Dong, X. Luo, K. S. Choi and X. Wu, “Heartbeat classification using disease-specific feature selection,” Computers in biology and medicine, vol. 46, pp. 79-89, 2014.
    [50] T. Mar, S. Zaunseder, J. P. Martínez, M. Llamedo and R. Poll, “Optimization of ECG classification by means of feature selection,” IEEE transactions on Biomedical Engineering, vol. 58, no. 8, pp. 2168-2177, 2011.
    口試委員
  • 俊侯良 - 召集委員
  • 劉志峰 - 委員
  • 林展霈 - 委員
  • 歐陽振森 - 委員
  • 李錫智 - 指導教授
  • 口試日期 2019-07-29 繳交日期 2019-09-03

    [回到前頁查詢結果 | 重新搜尋]


    如有任何問題請與論文審查小組聯繫