Automatic ECG arrhythmias classification scheme based on the conjoint use of the multi-layer perceptron neural network and a new improved metaheuristic approach

The authors have proposed a new automatic classification scheme based on the conjoint use of the multi-layer perceptron (MLP) neural network and an enhanced particle swarm optimisation (EPSO) algorithm for its training. In this work, six predominant categories of heartbeats from MIT-BIH database are considered, which are: normal, premature ventricular contraction, atrial premature contraction, right bundle branch block, left bundle branch block and paced beats. First, the authors have applied the standard particle swarm optimisation (PSO) algorithm to select the network structure for each features vector. Then, the relevant electrocardiogram (ECG) features to the studied arrhythmias were chosen, which suited to the optimised training performance of the classifier. The recognition performance of the proposed EPSO-MLP classification system is evaluated considering two different versions of the EPSO algorithm. In the first version (EPSOw), the inertia weight factor of the PSO algorithm is proposed to be a variable with iterations. However, two PSO parameters are taken to be variables in the second version of the improved learning algorithm (EPSOwc). The obtained experimental results prove the enhancement of the convergence ability of the MLP neural network and confirm the superiority of the proposed EPSO-MLP classification scheme on comparison with the other last published classification systems.

[1]  Kup-Sze Choi,et al.  Heartbeat classification using disease-specific feature selection , 2014, Comput. Biol. Medicine.

[2]  Fethi Bereksi-Reguig,et al.  Automatic Classification of Heartbeats Using Wavelet Neural Network , 2012, Journal of Medical Systems.

[3]  M. Arthanari,et al.  Classification of electrocardiogram signals with support vector machines and extreme learning machine , 2011, Neural Computing and Applications.

[4]  Rajarshi Gupta,et al.  Delineation of ECG characteristic features using multiresolution wavelet analysis method , 2012 .

[5]  K. Egiazarian,et al.  Comparative study of morphological and time-frequency ECG descriptors for heartbeat classification. , 2006, Medical engineering & physics.

[6]  Hazlie Mokhlis,et al.  Modified particle swarm optimization with time varying acceleration coefficients for economic load dispatch with generator constraints , 2014 .

[7]  Carsten Peterson,et al.  Clustering ECG complexes using Hermite functions and self-organizing maps , 2000, IEEE Trans. Biomed. Eng..

[8]  Messaoud Benidir,et al.  Multiresolution wavelet-based QRS complex detection algorithm suited to several abnormal morphologies , 2014, IET Signal Process..

[9]  Yu-Liang Hsu,et al.  ECG arrhythmia classification using a probabilistic neural network with a feature reduction method , 2013, Neurocomputing.

[10]  Di Wang,et al.  Automatic recognition of arrhythmia based on principal component analysis network and linear support vector machine , 2018, Comput. Biol. Medicine.

[11]  Juan Pablo Martínez,et al.  Heartbeat Classification Using Feature Selection Driven by Database Generalization Criteria , 2011, IEEE Transactions on Biomedical Engineering.

[12]  U. Rajendra Acharya,et al.  ECG beat classification using PCA, LDA, ICA and Discrete Wavelet Transform , 2013, Biomed. Signal Process. Control..

[13]  Philip de Chazal,et al.  Automatic classification of heartbeats using ECG morphology and heartbeat interval features , 2004, IEEE Transactions on Biomedical Engineering.

[14]  Michael R. Lyu,et al.  A hybrid particle swarm optimization-back-propagation algorithm for feedforward neural network training , 2007, Appl. Math. Comput..

[15]  Jian Li,et al.  Heartbeat classification using projected and dynamic features of ECG signal , 2017, Biomed. Signal Process. Control..

[16]  Elif Derya Übeyli,et al.  ECG beat classifier designed by combined neural network model , 2005, Pattern Recognit..

[17]  William Robson Schwartz,et al.  ECG-based heartbeat classification for arrhythmia detection: A survey , 2016, Comput. Methods Programs Biomed..

[18]  Eleonora Aronica,et al.  Gliomas and epilepsy: Insights from neuropathological studies in humans , 2015, SpringerPlus.

[19]  Yudong Zhang,et al.  Crop Classification by Forward Neural Network with Adaptive Chaotic Particle Swarm Optimization , 2011, Sensors.

[20]  Ahmed A. Ewees,et al.  Improving Twin Support Vector Machine Based on Hybrid Swarm Optimizer for Heartbeat Classification , 2018 .

[21]  Ivan W. Selesnick,et al.  Biomedical Signal Processing and Control Ecg Enhancement and Qrs Detection Based on Sparse Derivatives , 2022 .

[22]  Padmavathi Kora,et al.  Hybrid Firefly and Particle Swarm Optimization algorithm for the detection of Bundle Branch Block , 2016 .

[23]  Mohd Fadzli Mohd Salleh,et al.  Simultaneous Determination of Modulation Types and Signal-to-Noise Ratios Using Feature-Based Approach , 2018, IEEE Access.

[24]  Naomie Salim,et al.  Arrhythmia recognition and classification using combined linear and nonlinear features of ECG signals , 2016, Comput. Methods Programs Biomed..

[25]  José Ignacio Peláez,et al.  Automatic identification of characteristic points related to pathologies in electrocardiograms to design expert systems , 2019, Soft Comput..

[26]  Lifeng Xi,et al.  An Improved Particle Swarm Optimization for Evolving Feedforward Artificial Neural Networks , 2007, Neural Processing Letters.

[27]  Mehmet Korürek,et al.  ECG beat classification using particle swarm optimization and radial basis function neural network , 2010, Expert Syst. Appl..

[28]  Jooyoung Lee,et al.  Inverse Resolution Limit of Partition Density and Detecting Overlapping Communities by Link-Surprise , 2016, Scientific Reports.