The Application of Two Dimensional Principal Component Analysis in the Detection of Myocardial Infarction ECG Signals
GE Ding-fei1, XU Ai-qun2
1. School of Information and Electronic Engineering, Zhejiang University of Science and Technology, Hangzhou, Zhejiang 310012, China;
2. School of Mechanical and Automotive Engineering,Zhejiang University of Science and Technology, Hangzhou, Zhejiang 310012, China
Abstract：The joint energy percentage(EP) search method based on two dimensional principal component analysis is introduced to extract global features from 12-lead high resolution electrocardiogram(ECG) for the purpose of classification. Four types of classes are collected from PTB clinical diagnostic database, which corresponding patient’s statuses are health control, myocardial infarction(MI) in early stage, MI in acute stage and MI in recover stage, respectively. The experimental results show that the information can be fused efficiently using the proposed method, which are from 12-lead ECG and the details contained in high frequency components of ECG. The average classification accuracy can be increased by 10.43% compared with that of conventional principal component analysis. The ECG signal can be represented with lower dimensions compared with that of independent EP criterion, and an average classification accuracy of 99.46% can be achieved.
葛丁飞, 徐爱群. 二维主元分析法在心肌梗死心电信号检测中的应用[J]. 计量学报, 2014, 35(3): 252-257.
GE Ding-fei, XU Ai-qun. The Application of Two Dimensional Principal Component Analysis in the Detection of Myocardial Infarction ECG Signals. Acta Metrologica Sinica, 2014, 35(3): 252-257.
［1］ Dhruva N V, Abdelhadi S I, Anis A, et al. ST-segment analysis using wireless technology in actue myocardial infarction (STAT-Trial) ［J］.Journal of the American College of Cardiology, 2007, 50(6):509-13.
［2］ Huang C S. A vector cardiogram-based classification system for the detection of myocardial infarction ［C］// Proceeding in 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Boston, 2011.
［3］ Joo T H, Schmitt P W, Hampton D R, et al. Enhanced acute myocardial infarction detection algorithm using local and global signal morphology ［C］//Computer in Cardiology, Cleveland, OH,1998.
［4］ Edenbrandt H L, Ohllsson M. Detecting acute myocardial infarction in 12-lead ECG using Hermite expansions ［J］.Artificial Intelligence in Medicine,2004,32(2):127-136.
［5］ Abboud S, Zlochiver S. High-frequency QRS electrocardiogram for diagnosing and monitoring ischemic heart disease［J］.Journal of Electrocardiology,2006,39(1):82-86.
［6］ Mohamed H J, Noureddine Z. Face recognition system using relevance weighted two dimensional linear discriminant analysis algorithm［J］.Journal of Signal and Information Processing,2012,3(1):130-135.
［7］ Yang J, Zhang D, Frang A F, et al. Two-dimensional PCA: a new approach to appearance-based face representation and recognition［J］.IEEE Transaction Pattern Analysis and Machine Intelligence, 2004,26(1): 131-137.
［8］ Tompkins W J. Biomedical Digital Signal Processing ［M］. Englewood Cliffs, New Jersey: Prentice Hall,1993.
［9］ Anderson C W,Stolz E A,Shamssunder S.Multivariate autoregressive models for classification of spontaneous electroencephalographic signals during mental tasks［J］.IEEE Trans Biomed Eng,1998,45(3):277-286.
［10］ 葛丁飞, 孙丽慧, 文小军.基于Frank导联和Hermite展开的心肌梗死检测［J］.计量学报, 2009, 30(6): 551-554.
2014, Vol. 35 
