Dimension reduction and classification in ECG signal interpretation using FA & PCA: A comparison
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Electrocardiogram (ECG) is the electrical activity of the heart. It consists of P-wave, QRS complex and T-wave. Each beat shows electrical impulse propagation in the heart of the body. Due to higher dimension of ECG signal, its analysis needs dimension reduction tech- niques for pre-processing of the data. In this paper, authors proposed Factor Analysis (FA) and Principal Component Analysis (PCA) for re- ducing dimension of the ECG signal. The choice of its usage depends on problem which is clearly known and speci ed. Physionet ECG data- base (MIT-BIH long term ECG database, Ventricular Tachyarrhythmia, AHA(American Heart Association) Database) and real time ECG data- base has been used in this paper for checking detector performance using FA and PCA. PCA gave better results than FA. SNR (Signal to Noise Ratio) is also checked and calculated that is 93.25 dB and 91.25 dB for PCA and FA respectively. Di erent work proposed by di erent authors on ECG signal classification has been compared.
Abstract
Electrocardiogram (ECG) is the electrical activity of the heart. It consists of P-wave, QRS complex and T-wave. Each beat shows electrical impulse propagation in the heart of the body. Due to higher dimension of ECG signal, its analysis needs dimension reduction tech- niques for pre-processing of the data. In this paper, authors proposed Factor Analysis (FA) and Principal Component Analysis (PCA) for re- ducing dimension of the ECG signal. The choice of its usage depends on problem which is clearly known and speci ed. Physionet ECG data- base (MIT-BIH long term ECG database, Ventricular Tachyarrhythmia, AHA(American Heart Association) Database) and real time ECG data- base has been used in this paper for checking detector performance using FA and PCA. PCA gave better results than FA. SNR (Signal to Noise Ratio) is also checked and calculated that is 93.25 dB and 91.25 dB for PCA and FA respectively. Di erent work proposed by di erent authors on ECG signal classification has been compared.
참고문헌 (15)
- C.Bakir, "Classzfication of ECG Signals By the Neighborhood Feature Extraction Method", Balkan Journal of Electrical and Computer Engineering, 3 (2015)
- Z. Zidelmal, "QRS detection based on wavelet coefficients", computer methods and programs in biomedicine, 107, pp.490496~ (2012)
- I. Saini, "QRS detection using K-Nearest Neighbor algorithm (KNN) and evaluation on standard ECG databases", Journal of Advanced Research, (4), pp.331344~ (2013)
- N. Maglaveras, "ECG pattern recognition and classification using non-linear transformations and neural networks, A review", International Journal of Medical Informatics, 52, pp.191~208 (1998)[DOI]
- M. Zivanovic et.al, "Simultaneous powerline interference and baseline wander removalfrom ECG and EMG signals by sinusoidal modelling", Medical Engineering and Physics, 35, pp.1431~1441 (2013)[DOI]
- "MIT-BIH arrhythmia database"[URL]
- B.T Krishna, "Electrocardiogram Signal and Linear Time-Frequency Transforms", J.Inst.Eng. India Ser.B., 95(4), pp.377~382 (2014)[DOI]
- M.P.S. Chawla et. ai, "A new statistical PCAICA algorithm for-location of R-peaks in ECG", International Journal of Cardiology, 129, pp.146148~ (2008)
- F. Castells, "Principal Component Analysis in ECG Signal Processing", EURASIP Journal on Advances in Signal Processing, 2007 (2007), Hindawi Publishing Corporation
- L.C. Lee et al., "Comparison of Several Variants of Principal Component Analysis (PCA) on Forensic Analysis of Paper Based on IR Spectrum", AlP Conference Proceedings, 1750, pp.060012~ (2016)
- R.J. Martis, "Current methods in electrocar-diogram characterization", Computers in Biology and Medicine, 48, pp.133149~ (2014)
- R.J. Martis, "Characterization of ECG beats from cardiac ar-rhythmia using discrete cosine transform in PCA framework", Knowledge-Based Systems, 45, pp.7682~ (2013)
- R.J. Martis, "Application of principal component analysis to ECG signals for automated diagnosis of cardiac health", Expert Systems with Applications, 39, pp.1179211800~ (2012)
- G. Guo, "KNN Model-Based Approach in Classification", ~
- N. Uchaipichat et.al, "Development of QRS Detection using Short-time Fourier Transform based Technique", IJCA Special Issue on Computer Aided Soft Computing Techniques for Imaging and Biomedical Applications, pp.7~10 (2010), CASCT