以斜率為基礎建立偵測心電圖R波之研究

The Detection of the R Wave of Electrocardiogram Based on Concept of Slope.

陳建德(JIAN DE CHEN)

指導教授 ： 江行全

元智大學/工程學院/工業工程與管理學系/碩士(2006年)

https://doi.org/10.6838/YZU.2006.00215

摘要

醫師與醫護人員通常使用心電圖(Electrocardiogram, ECG)來判讀心臟異常訊息，以及如何正確地辨別各類心電圖，但這些判讀工作需要耗費大量的人工資源與時間，常因過於疲勞或是個人主觀因素造成誤判的情形發生，因此本研究的目的是如何正確地判斷出心電圖中最重要的特徵-R波。對於一般較規律且雜訊較少的心電圖而言，以斜率理論為主的辨識方法均有不錯的辨識效果，但對於雜訊影響較大的心電圖，Q-R區間會有多次轉折點的情況發生，導致無法正確判斷，因此本研究針對So and Chan方法加入連續遞增數(Continuous runs)的觀念改進其缺失，並以Massachusetts Institute of Technology-Beth Israel Hospital (MIT/BIH)為訓練資料，同時用實際醫務心電圖資料做驗證及測試，且與相關文獻中所提及的6種方法進行比較，結果發現本研究提出的修正法比原始方法效果更好，在訓練與測試資料中均有效地降低雜訊所造成的誤判，R波平均準確率從原始So and Chan方法所得89.85%提升至修正法的98.35%，並能準確地偵測出R波位置所在。

關鍵字

心電圖 (ECG) ； QRS波 ； 連續遞增數(Continuous runs) ； MIT/BIH心律不整資料庫

並列摘要

Doctors and medical personnel usually use electrocardiogram (ECG) signal to diagnose the abnormal information of the heart and to classify each kind of ECG. The diagnosis process needs a lot of human resources and time, and it often causes incorrect judgment because of fatigue or personal bias. For this reason, the purpose of this study is how to correctly judge the characteristic R wave of ECG. The methods based on slope concept presented good performances to regular and less noise of on ECG pattern. If there were significant noises on ECG data, it produced several turned points in the region between Q and R wave, and that couldn’t diagnose properly. The research revised So and Chan method and combined with the concept of continuous runs to improve the weakness on QRS complex. The diagnosis technique is trained by Massachusetts Institute of Technology-Beth Israel Hospital (MIT-BIH) arrhythmia database and verified by real medical ECG data. The results showed that the revised So and Chan method could obtain better performances than the other traditional methods mentioned literature. The proposed method produces 98.35% accuracy to detect R wave locations on ECG data, and improves shortcoming of So and Chan method effectively.

並列關鍵字

Electrocardiogram (ECG) ； continuous runs ； QRS complex ； MIT-BIH database

參考文獻

7. 陳彥宏，”判別分析在心電圖上的應用”， 國立中山大學應用數學研究所，碩士論文，民國86年。

Nagle,”A comparison of the noise sensitivity of nine QRS detection algorithms”,

15. J. Lee, K. Jeong, J. Yoon; M. Lee, ”A Simple Real-Time QRS Detection Algorithm,” 18th Annual Intern. Conference of the IEEE Engineering in Medicine and Biology Society, Vol. 4, pp. 1396-1398, Nov. 1996.

16. J. Neves Rodrigues, V. Owall, and L. Sornmo, “QRS detection for pacemakers in a noisy environment using a time lagged artificial neural network,” IEEE International Symposium vol. 3 pp. 596-599, May. 2001

17. K. F. Tan, K. L. Chan, K. Choi, “Detection of the QRS complex, P wave and T wave in electrocardiogram,” Processing of 2000 IEE Conference on Advances in Medical Signal and Information Processing, pp. 41-47, 2000.