ECG fiducial point extraction using switching Kalman filter

Abstract

In this paper, we propose a novel method for extracting fiducial points (FPs) of the beats in electrocardiogram (ECG) signals using switching Kalman filter (SKF). In this method, according to McSharry's model, ECG waveforms (P-wave, QRS complex and T-wave) are modeled with Gaussian functions and ECG baselines are modeled with first order auto regressive models. In the proposed method, a discrete state variable called "switch" is considered that affects only the observation equations. We denote a mode as a specific observation equation and switch changes between 7 modes and corresponds to different segments of an ECG beat. At each time instant, the probability of each mode is calculated and compared among two consecutive modes and a path is estimated, which shows the relation of each part of the ECG signal to the mode with the maximum probability. ECG FPs are found from the estimated path. For performance evaluation, the Physionet QT database is used and the proposed method is compared with methods based on wavelet transform, partially collapsed Gibbs sampler (PCGS) and extended Kalman filter. For our proposed method, the mean error and the root mean square error across all FPs are 2 ms (i.e. less than one sample) and 14 ms, respectively. These errors are significantly smaller than those obtained using other methods. The proposed method achieves lesser RMSE and smaller variability with respect to others.

Electrocardiogram Fiducial Point Detector Using a Bilateral Filter and Symmetrical Point-Filter Structure

“…Therefore, the detection definition for FPs is not simple. Nevertheless, many different PQRST complex detection approaches based on (1) filtering techniques, such as adaptive filtering [32], low-pass differentiation (LPD) [33], nested median filtering [34], extended Kalman filtering [35,36], and switching Kalman filter [37]; (2) transform coefficient techniques, such as discrete Fourier transform (DFT) [38], discrete cosine transform (DCT) [39], wavelet [40][41][42], and hybrid feature extraction algorithm (HFEA) [42]; and (3) inference model techniques, such as Bayesian [43][44][45], hidden Markov model [46,47], and partially collapsed Gibbs Sampler (PCGS) [44,45], can be found in the literature. These methods exhibit excellent performance in identifying ECG FPs.…”

A Framework for Segmentation and Classification of Arrhythmia Using Novel Bidirectional LSTM Network

“…The timefrequency approach has proven highly effective in extracting frequency at a finer resolution [14], [15] [16]. SVM, radial basis function NN-based classifier [17], multilayer perceptron and various search algorithms have enhanced the performance of classifiers [18], [19]. Many of the methods discussed in the given literature employ hardcoded features for signal processing, segmentation and detection, which eventually leads to high false positives and consequently to misdiagnosis.…”

Clinical interpretations of the effectiveness of changes in body position during aerobic fitness after neurologic injury

“…Systematic literature reviews over the past three decades of QRS detection methods are conducted by Elgendi et al (2014); Kohler et al (2002); and Martis et al (2013). The typical proposed methods can be categorized into several types: (i) derivatives and digital filtering methods (Castells-Rufas and Carrabina, 2015;Chatterjee et al, 2012;Hamilton and Tompkins, 1986;Karimipour and Homaeinezhad, 2014;Ning and Selesnick, 2013;Pan and Tompkins, 1985;Phukpattaranont, 2015), (ii) transformation-based methods (Bono et al, 2014;Ieong et al, 2014;Madeiro et al, 2013;Martínez et al, 2010;Merah et al, 2015;Pal and Mitra, 2012;Yochum et al, 2016;Zhu and Dong, 2013;Zidelmal et al, 2014), (iii) machine learning methods (Akhbari et al, 2018Lin et al, 2010) and (iv) other methods (curve fitting (Tafreshi et al, 2014), mathematical modeling (Madeiro et al, 2013), template matching (Bashir et al, 2014;Chen and Chuang, 2017), correlation analysis (Homaeinezhad et al, 2014;Karimipour and Homaeinezhad, 2014), point process tracking (Citi et al, 2012), power analysis (Kim and Shin, 2016), and masking & amplitude analysis (Chen and Chuang, 2017).…”