VFPred: A fusion of signal processing and machine learning techniques in detecting ventricular fibrillation from ECG signals

Abstract

Ventricular Fibrillation (VF), one of the most dangerous arrhythmias, is responsible for sudden cardiac arrests. Thus, various algorithms have been developed to predict VF from Electrocardiogram (ECG), which is a binary classification problem. In the literature, we find a number of algorithms based on signal processing, where, after some robust mathematical operations the decision is given based on a predefined threshold over a single value. On the other hand, some machine learning based algorithms are also reported in the literature; however, these algorithms merely combine some parameters and make a prediction using those as features. Both the approaches have their perks and pitfalls; thus our motivation was to coalesce them to get the best out of the both worlds. Hence we have developed, VFPred that, in addition to employing a signal processing pipeline, namely, Empirical Mode Decomposition and Discrete Fourier Transform for useful feature extraction, uses a Support Vector Machine for efficient classification. VFPred turns out to be a robust algorithm as it is able to successfully segregate the two classes with equal confidence (Sensitivity = 99.99%, Specificity = 98.40%) even from a short signal of 5 seconds long, whereas existing works though requires longer signals, flourishes in one but fails in the other.

Ventricular Fibrillation and Tachycardia Detection Using Features Derived from Topological Data Analysis

“…However, they achieved an extremely low value of sensitivity (Sen = 56.44%). Finally, in 2019 Ibtehaz et al [73] got the highest results in this group, using a scheme of incorporating Empirical Mode Decomposition (EMD) and SVM classifiers (Sen = 99.99%, Spe = 98.40%, Acc = 99.19%) for the classification of VF and non-VF episodes.…”

An Improved Empirical Mode Decomposition Based on Local Integral Mean and Its Application in Signal Processing

“…Figure 21 shows the original ECG100 (numbered as 100 in the database), MA, BW waveforms [52], and the noised ECG100 waveform formed by superimposing these three signals, and Figure 22 shows all IMFs of the noised ECG100 decomposed by EMD, EEMD, CEEMDAN, and IMMD. In ECG reconstruction, IMF 1 Mathematical Problems in Engineering contains a lot of high-frequency noise, which is generally recognized as MA, so it is often removed [53][54][55][56]. For the remaining IMFs, the QRS characteristic wave is used to identify the mode components of ECG.…”

A Lightweight CNN Model for Detecting Respiratory Diseases from Lung Auscultation Sounds using EMD-CWT-based Hybrid Scalogram

“…Table I represents data distribution at several levels of processing corresponding to the disease classes considered for this study. C. Feature extraction 1) Empirical Mode Decomposition (EMD): EMD is a powerful self-adaptive signal decomposition method especially in the time scale and energy distribution aspects and highly suitable for analysis and processing of non-linear and nonstationary signals such as lung sounds and heart sounds [54]. It decomposes a given signal x(t) into a finite set (N) of intrinsic mode functions, IMF 1 (t), IMF 2 (t), .…”

“…An IMF is a simple oscillatory function with the equal number of extrema and zero crossings and its envelopes must be symmetrical with respect to zero. Thus, the EMD detrends a signal and elicits underlying spectral patterns [54].…”