結合隱藏式馬可夫模型與高斯混合模型於12導程心電圖之冠心症疾病辨識

A Hybrid System with Hidden Markov Models and Gaussian Mixture Models for Myocardial Infarction Classification with 12-Lead ECGs

周晏弘(Yen-Hung Chou)

指導教授 ： 張百棧

元智大學/資訊學院/資訊管理學系/碩士(2009年)

https://doi.org/10.6838/YZU.2009.00044

摘要

此論文中，將針對12導程心電資料進行冠心症疾病的模型建立，在臨床醫學中，透過心電圖，醫師可以第一時間發現病人所患的心臟疾病為何並給予正確的治療。目前心電儀器所提供的預測準確率不高，還有醫師經驗及個人主觀，心電圖資料格式的不統一，所以在心電醫療研究這方面一直有所瓶頸。而透過之前的研究，已經將心電圖病歷進行解碼，所以可以自行開發相關醫療應用。此研究將利用隱藏式馬可夫模型針對冠心症疾病，進行疾病預測的模型建立，透過此法，我們可以找出隱藏在心電圖中冠心症的特徵值。本實驗將針對冠心症中的心肌梗塞作為研究目標，資料數量為真實患有心肌梗塞的病歷500份，而無罹患心臟疾病的病歷515份。首先透過患有疾病的心電資料，將建立從Lead V1到Lead V4導程的隱藏式馬可夫模型，接著將訓練好的模型，進行測試資料的相似值比對。將患有心肌梗塞疾病與無患心臟疾病測試資料的相似值資料，放入高斯混合模型當中分類，再比較其準確率，來證實我們的方法具有醫學上的幫助，可以幫助醫生在診斷時，提供更佳的診斷意見，縮短搶救的時間。而經過辨識之後，我們得到最高準確率為83％，因真實資料變異大，所以此研究準確率不及傳統研究高，但跟傳統的研究相比，卻具有更高的實用價值，在未來的研究，可利用別的方法加入，提高準確率，給予醫學界更大的幫助。

關鍵字

隱藏式馬可夫模型 ； 高斯混合模型 ； 冠心症疾病辨識

並列摘要

This study presented a new diagnosis system with integrating 12-lead ECG data into a density model for increasing accuracy rate and flexibility of diseases detection. A hybrid system with HMMs and GMMs was employed for data classification. For myocardial infarction, data of lead-V1, V2, V3 and V4 were selected and HMMs were used not only to find the ECG segmentations but also to calculate the log-likelihood value which was treated as statistical feature data of each heartbeat’s ECG complex. The 4-dimension feature vector was clustered by GMMs and different numbers of distribution (disease and normal data) were examined in experiment. The main idea in this study relied on the multiple ECG channels which could be combined. There were total 1015 samples of heartbeats from clinical data, including 500 data with myocardial infarction and 515 normal data. The accuracy of this diagnosis system achieved 83%.

並列關鍵字

Hidden Markov Models ； Gaussian Mixture Models ； Myocardial Infarction Classification

參考文獻

[2] “American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care - Part 8: Stabilization of the Patient With Acute Coronary Syndromes,” December. 2005.

[3] K. Minami, H. Nakajima, T. Toyoshima, “Real-time discrimination of ventricular tachyarrhythmia with Fourier-transform neural network,” Biomedical Engineering, IEEE Transactions on, vol. 46, 1999, P. 179-185.

[4] Cuiwei Li, Chongxun Zheng, Changfeng Tai, “Detection of ECG characteristic points using wavelet transforms,” Biomedical Engineering, IEEE Transactions on, vol. 42, 1995, P. 21-28.

[5] D. Coast, R. Stern, G. Cano, S. Briller, “An approach to cardiac arrhythmia analysis using hidden Markov models,” Biomedical Engineering, IEEE Transactions on, vol. 37, 1990, P. 826-836.

[6] L. Rabiner, B. Juang, “An introduction to hidden Markov models,” ASSP Magazine, IEEE, vol. 3, 1986, P. 4-16.