A novel automated tower graph based ECG signal classification method with hexadecimal local adaptive binary pattern and deep learning

Abstract

Electrocardiography (ECG) signal recognition is one of the popular research topics for machine learning. In this paper, a novel transformation called tower graph transformation is proposed to classify ECG signals with high accuracy rates. It employs a tower graph, which uses minimum, maximum and average pooling methods altogether to generate novel signals for the feature extraction. In order to extract meaningful features, we presented a novel one-dimensional hexadecimal pattern. To select distinctive and informative features, an iterative ReliefF and Neighborhood Component Analysis (NCA) based feature selection is utilized. By using these methods, a novel ECG signal classification approach is presented. In the preprocessing phase, tower graph-based pooling transformation is applied to each signal. The proposed one-dimensional hexadecimal adaptive pattern extracts 1536 features from each node of the tower graph. The extracted features are fused and 15,360 features are obtained and the most discriminative 142 features are selected by the ReliefF and iterative NCA (RFINCA) feature selection approach. These selected features are used as an input to the artificial neural network and deep neural network and 95.70% and 97.10% classification accuracy was obtained respectively. These results demonstrated the success of the proposed tower graph-based method.

ECG signal processing by feature selection with classification based on machine learning architectures in cardiovascular disease detection

“…Because a special instrument was used for function extraction, the approach's main disadvantage was the lack of automated recognition, which limited its benefits. Heart rate variability (HRV) has been changed into the machine by work [10] so that it can be used to identify people. Author [11] has created a collection of descriptors to characterise an ECG graph.…”

“…where 𝑀 = 1 1 is margin solution of min 1≤𝑘≤𝑛 𝑦 𝑘 (𝜃 ⊤ 𝑥 𝑘 + 𝑏). The following optimization issue has a solution in the form of the hyperplane by eqn (10):…”

Arrhythmia Disease Diagnosis Based on ECG Time–Frequency Domain Fusion and Convolutional Neural Network

“…Jikuo et al [24] proposed a novel convolutional neural network with a non-local convolutional block attention module (NCBAM) for ECG signal classification of single heartbeat cycles. Subasi et al [25] Eigenvalue extraction of the ECG signal is to extract some of the key information of the ECG signal that contains the heart activity, so that the ECG signal can be represented in a lower dimension. This information can be used to diagnose ECG signals with high accuracy with less information.…”

“…Jikuo et al [24] proposed a novel convolutional neural network with a non-local convolutional block attention module (NCBAM) for ECG signal classification of single heartbeat cycles. Subasi et al [25] used an iterative relief and neighborhood component analysis (NCA) based feature selection approach for ECG signals. The final features were fed into a deep neural network (DNN) to obtain good diagnostic results.…”

A fully-automated paper ECG digitisation algorithm using deep learning

“…Our tool can facilitate the rapid and automated digitisation of large repositories of paper ECGs to allow them to be used for deep learning projects.There has been growing interest in applying machine learning to electrocardiograms (ECG). For example, variations of wavelet analysis and local binary patterns were used for extracting features from ECG, then support vector machine (SVM), k-nearest neighbour (kNN), and state-of-the-art deep neural networks were explored for arrhythmia diagnosis [1][2][3][4][5] . Convolutional neural networks (CNN) have also been used to predict the likelihood of paroxysmal atrial fibrillation (AF) from sinus rhythm ECGs 6-8 , screening left ventricular systolic dysfunction to identify incident heart failure 9-14 , screening hypertrophic cardiomyopathy [15][16][17] , and early diagnosis of valvular diseases such as aortic stenosis and mitral regurgitation [18][19][20] .…”