Electrocardiogram signal analysis for R-peak detection and denoising with hybrid linearization and principal component analysis

Abstract

Abstract:In the areas of biomedical and healthcare, electrocardiogram (ECG) signal analysis is one of the major aspects of research. The accuracy in the detection of subtle characteristic features in ECG is of great significance. This paper deals with an algorithm based on hybrid linearization and principal component analysis for ECG signal denoising and R-peak detection. The ECG data have been taken from the MIT-BIH Arrhythmia Database for performance evaluation.The signal is denoised by applying the hybrid linearization method, which is an arrangement of the extended Kalman filter along with discrete wavelet transform, and then principal component analysis is employed to detect R waves and the QRS complex. The reported work has been implemented in the MATLAB environment for 25 different ECG records yielding 99.90% sensitivity, 99.97% positive predictivity, and a detection error rate of 0.120%. The achieved performance outperforms the recent research done in the area of interest.

An Adaptive ECG Noise Removal Process Based on Empirical Mode Decomposition (EMD)

“…Several studies have contributed many reviews to propose the formulation of techniques for computerised ECG denoising. Such developed methods are primarily based on deep learning techniques [14,15], deep recurrent neural networks (RNN) [16], filter banks [17], time-frequency techniques [18][19][20], discrete wavelet transform (DWT) filtering [21,22], empirical mode decomposition (EMD) [23][24][25][26], impulse response (FIR) filter [17,27], nonlocal mean (NLM) filter [28], and principle component analysis (PCA) [29,30].…”

“…Systems based on neural networks [ 14 , 15 ] involve additional training stage and representative signals, hence they are not appropriate for real-time requests. In PCA methods [ 29 , 30 ], the resulting numerical model is greatly complex to a small variation in the waves or even the noises. The efficiency of a NLM filter [ 28 ] is dependent on the accurate choice of structure bandwidth that can be computed from the artefacts' ST (standard deviation).…”