Classification of Atrial Fibrillation Recurrence Based on a Convolution Neural Network With SVM Architecture

Abstract

Although radio frequency ablation is the most effective treatment for atrial fibrillation (AF), there is still a high recurrence rate. The purpose of this paper was to initially assess the probability of the recurrence of AF based on the preoperative body surface potential mapping (BSPM) signals, in other words, to predict the efficiency of ablation and assist physicians in developing more effective treatment options. At present, deep learning methods based on convolutional neural networks (CNNs) do not require complex mathematical abstractions or manual interventions; thus, higher computation efficiency can be obtained in such research. However, the use of the fully connected multi-layer perceptron (MLP) algorithms has shown low classification performance. This paper proposes an improved CNN algorithm (CNN-SVM method) for the recurrence classification in AF patients by combining with the support vector machine (SVM) architecture. The algorithm is validated on the preoperative AF signals of 14 patients for classification. All postoperative patients are followed up for one year; ten of them remain in sinus rhythm, whereas the other four turn back to AF. The ECG data for these patients are obtained through the 128-Lead BSPM system. The results show that the proposed CNN-SVM method can automatically extract the characteristic information through the CNN network. The constructed model ultimately achieved an accuracy of 96%, a sensitivity of 88%, and a specificity of 96%. It is concluded that the CNN-SVM method solves the drawbacks of MLP only for separating linear data. It improves the overall performance of AF recurrence classification, thereby providing a valuable reference for doctors to develop personalized treatment plans. INDEX TERMS Classification of AF recurrence, deep learning, body surface potential mapping, support vector machines. The associate editor coordinating the review of this manuscript and approving it for publication was Yonghong Peng.

A Hybrid Deep CNN Model for Abnormal Arrhythmia Detection Based on Cardiac ECG Signal

“…Recently, 2D CNN models have been employed to categorize the input ECG signals into their respective classes. The input 1D ECGs are transformed into 2D before the feature extraction process [ 16 ]. The 2D CNN models offer more distinctiveness and robustness towards noise in the input signals.…”

“…ECG signals contain no less than two critical pieces of statistics, including correlated to biomedicine’s healthiness [ 3 , 4 , 5 ] and associated with personal credentials or biometrics [ 6 , 7 , 8 ]. As a result of its easiness, several ECG categorizations processes have been established, counting manuals methods [ 9 , 10 ] and machine learning approaches [ 11 , 12 , 13 , 14 , 15 , 16 ]. The manual process is complicated.…”

Deep Learning Algorithm Classifies Heartbeat Events Based on Electrocardiogram Signals

“…With more and more medical data waiting to be mined, improving performance with acceptable computational complexity for algorithms and models is urgent. Li et al (2019) propose combining CNN and SVM to build an AF recognition model, which improves model performance from 93 to 96% compared with the CNN model. Swapna et al (2018) have attempted to compare CNN net with a combination of CNN and LSTM in processing a diabetes database.…”

“…With the development of machine and deep learning technologies, more ECG category classification models have been proposed, enabling the automatic diagnosis of different heart diseases. Of these technologies, the support vector machine (SVM) approach ( Li et al, 2019 ), random forest method ( Yang et al, 2020 ), autoregressive modeling ( Ge et al, 2002 ), artificial ( Xu et al, 2015 ), and convolutional neural networks (CNNs; Kiranyaz et al, 2016 ; Hannun et al, 2019 ), and long short-term memory (LSTM) have been mainly used to establish ECG classification models ( de Chazal et al, 2004 ; Asl et al, 2008 ; Acharya et al, 2017 ; Yildirim, 2018 ). To date, several heartbeat abnormalities have been frequently studied, such as AF, premature ventricular contraction (PVC), paced beat, and left or right bundle branch block (LBBB or RBBB, respectively; Raj and Ray, 2018 ).…”

An Efficient Algorithm for Cardiac Arrhythmia Classification Using Ensemble of Depthwise Separable Convolutional Neural Networks

“…There are at least two types of important information contained in the ECG signal, including those related to health or biomedical [2][3][4] and those related to the person identification or biometrics [5][6][7]. Due to its convenience, many ECG classification algorithms have been developed, including handcraft [4,8,9] and machine learning [10][11][12][13][14][15] methods. The handcraft method is rather difficult to utilize on non-stationary signals, such as ECG, while machine learning methods normally require high computational resources.…”