Arrhythmia Detection in ECG Signal Using Fractional Wavelet Transform with Principal Component Analysis
- PDF / 1,309,439 Bytes
- 11 Pages / 595.276 x 790.866 pts Page_size
- 36 Downloads / 236 Views
ORIGINAL CONTRIBUTION
Arrhythmia Detection in ECG Signal Using Fractional Wavelet Transform with Principal Component Analysis Varun Gupta1 • Monika Mittal2
Received: 28 November 2018 / Accepted: 3 September 2020 Ó The Institution of Engineers (India) 2020
Abstract Any significant alteration in the Electro-CardioGram (ECG) signal wave components (P-QRS-T) for a time duration is detected as arrhythmia. In this paper, a novel fractional wavelet transform (FrWT) is used as a preprocessing tool. FrWT describes the given signal in time–frequency fractional domain using fractional Fourier transform and its denoising using wavelet transform. Because of this novel and intriguing property, it is broadly utilized as a noise removal tool in the fractional domain along with multiresolution analysis. Next, features are extracted using Yule–Walker autoregressive modeling. Dimensionality of the extracted features is to be reduced for proper detection of different types of arrhythmias. Principal component analysis has been applied for arrhythmia detection using variance estimation. The proposed method is evaluated on the basis of various performance parameters such as output SNR, mean squared error (MSE) and detection accuracy (DEAcc ). An output SNR of 33.41 dB, MSE of 0.1689% and Acc of 99.94% for realtime ECG database and output SNR of 25.25 dB, MSE of 0.1656%, DEAcc of 99.89% for MIT-BIH Arrhythmia database are obtained. Keywords Electro-Cardio-Gram Fractional wavelet transform Multiresolution analysis Yule–Walker autoregressive modeling
& Varun Gupta [email protected]; [email protected] 1
KIET Group of Institutions, Delhi-NCR, Ghaziabad, UP 201206, India
2
National Institute of Technology, Kurukshetra, Haryana 136119, India
Introduction Detection of R-peak (QRS-complex) in the Electro-CardioGram (ECG) signal indicates correct heart condition of the subject by non-invasive heart rate variability (HRV) measurements [1–5]. In the existing literature, different approaches [6–9] have been proposed for preprocessing of ECG signals, but they failed to have wide acceptance due to introducing fluctuations in the QRS-complex. Choice of the ECG signal preprocessing technique is critical as it greatly affects the final outcome [10–14]. Therefore, in this paper, fractional wavelet transform (FrWT) has been proposed as a new multiresolution preprocessing tool in the fractional domain, as it possesses good mathematical attributes of both the wavelet transform (WT) and the fractional Fourier transform (FrFT) along with its own traits [15]. In this research work, FrWT, Yule–Walker autoregressive modeling (YWARM) and principal component analysis (PCA) have been adopted to detect several types of arrhythmias. Selection of proper model order is essential in YWARM to optimize number of equations involved, thereby reducing the complexity [16, 17]. PCA helps in detecting a R-peak by estimating eigenvalues and eigenvectors of the new orthogonal basis [18]. Finally, maximum values of the estimated variance help in the detection of
Data Loading...
