透過您的圖書館登入
IP:44.192.132.66
  • 學位論文

應用智慧型計算在12導程心電圖於心臟疾病之辨識

Cardiac Disease Classification with Clinical 12-Lead ECG by Using Computational Intelligence Approach

指導教授 : 張百棧

摘要


在臨床病例醫學中,心電圖在心肺相關疾病上為一最為廣泛應用的非侵入式診斷工具。心電圖監測病人的心跳狀況,並對於心房與心室的活動給予精準且重要的資訊。在處理並瞭解心電圖波形的過程,關鍵在於其在時間上形態轉換之處理。臨床上,12導程心電圖在多數的醫院中均已有所應用,其對於心臟疾病提供了更多詳盡的波形資料。本研究將著重於利用不同的智慧型技術,透過分析12導程心電圖之形態,從中擷取心電訊號之特徵,並藉以判定該病例所可能之疾病。本研究中,使用心房顫動及心肌梗塞做為本研究之判定病例。於心房顫動病例中,本研究使用盲訊號分離技術,從12導程心電圖中分離出可能之顫動訊號,並將其視為用以判別病例之特徵;於心肌梗塞病例中,本研究則使用隱藏式馬可夫模型計算在不同導程中,該心跳波形可能之發生機率,本研究並使用主成份分析及多項式近似法擷取ST片段中之特徵資訊。此三類數學模型均使用於12 導程心電圖中計算出可用於判斷心肌梗塞之特徵選取,其中主成分分析對於單一心跳之診斷率較高而多項式近似法則對於病例之診斷較為有效。

並列摘要


In clinical medicine, Electrocardiogram (ECG) is one of the most widely used non-invasive diagnostic tools for cardiopulmonary diseases. ECG monitors the patients’ heart-beat and clinically gives accurate and important information about the activities of atrium and ventricle. The key in treating ECG complex is using the morphology in time detection. Clinical 12-lead ECG data is now available in most hospitals and it includes more detailed information about cardiac disease. This research focuses on how to extract the features in ECG data by analyzing the morphological characteristics through different approaches. Atrial Fibrillation (AF) and Myocardial Infarction (MI) are used to verify the applied computational intelligence approaches. For AF classification, Blind Source Separation (BSS) is adopted to separate the estimated fibrillation sources from 12-lead ECG. The extracted sources can be regarded as the features of AF and used to classify the input cases. Second, for MI classification, three mathematical models are applied to calculate the feature vector of MI cases. Hidden Markov Models (HMMs) are used to calculate the likelihood value for each lead; Principal Component Analysis (PCA) is adopted to find the main components of ST segment and polynomial approximation is applied to calculate the coefficients of fitted polynomial formula of ST segment. The calculated coefficients can be used as input feature vector for classifier. In experimental result with SVM classifier, PCA is better for beat classification while polynomial approximation can give the higher accuracy for case classification.

參考文獻


[46] S. Krimi, K. Ouni and N. Ellouze, "An approach combining Wavelet Transform and Hidden Markov Models for ECG segmentation," in 2008 3rd International Conference on Information and Communication Technologies: From Theory to Applications, ICTTA 2008.
[18] V. Fuster, L. E. Ryden, D. S. Cannom, H. J. Crijns, A. B. Curtis, K. A. Ellenbogen, J. L. Halperin, J. Y. Le Heuzey, G. N. Kay, J. E. Lowe, S. B. Olsson, E. N. Prystowsky, J. L. Tamargo, S. Wann, S. C. Smith Jr, A. K. Jacobs, C. D. Adams, J. L. Anderson, E. M. Antman, S. A. Hunt, R. Nishimura, J. P. Ornato, R. L. Page, B. Riegel, S. G. Priori, J. J. Blanc, A. Budaj, A. J. Camm, V. Dean, J. W. Deckers, C. Despres, K. Dickstein, J. Lekakis, K. McGregor, M. Metra, J. Morais, A. Osterspey, and J. L. Zamorano, "ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation: A report of the American College of Cardiology/American Heart Association Task Force on practice guidelines and the European Society of Cardiology Committee for practice guidelines (Writing committee to revise the 2001 guidelines for the management of patients with atrial fibrillation)," Circulation, vol. 114, pp. e257-e354, 2006.
[2] P. Trahanias and E. Skordalakis, "Syntactic pattern recognition of the ECG," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 12, pp. 648-657, 1990.
[3] N. Maglaveras, T. Stamkopoulos, K. Diamantaras, C. Pappas and M. Strintzis, "ECG pattern recognition and classification using non-linear transformations and neural networks: A review," International Journal of Medical Informatics, vol. 52, pp. 191-208, 1998.
[4] A. Gacek and W. Pedrycz, "A genetic segmentation of ECG signals," IEEE Transactions on Biomedical Engineering, vol. 50, pp. 1203-1208, 2003.

被引用紀錄


張冠群(2012)。醫療產業導入雲端服務關鍵成功因素之研究〔碩士論文,國立臺北科技大學〕。華藝線上圖書館。https://doi.org/10.6841%2fNTUT.2012.00603
李秋璇(2010)。應用網路教學於護理系學生暴力處置課程知識成效之探討〔碩士論文,中山醫學大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0003-1909201015102900

延伸閱讀