The nonlinear time-series analysis method, based on the recurrence plot theory, has received great attention from researchers and has been successfully used in multiple fields. However, traditional recurrence plots that use Heaviside step functions to determine the recursive behavior of a point in the phase space have two problems: (1) Heaviside step functions produce a rigid boundary, resulting in information loss; and (2) the selection of the critical distance, ε, is crucial; if the selection is inappropriate, it will result in a low-dimensional dynamics error, and as of now, there exists no unified method for selecting this parameter. With regard to the problems described above, the novelty of this article lies in the following: (1) when determining the state-phase point recursiveness, a Gaussian function is used to replace the Heaviside function, thereby solving the rigidity and binary value problems of the recursive analysis results caused by the Heaviside step function; and (2) texture analysis is performed on a recurrence plot, new ways of studying complex system dynamics features are proposed, and a system of complex system dynamic-like measurement methods is built.

/pdf/research-on-recurrence-plot-feature-quantization-method-m6ujwte2.pdf

Research on Recurrence Plot Feature Quantization Method Based on Image Texture Analysis

Deep learning (DL) has been introduced in automatic heart-abnormality classification using ECG signals, while its application in practical medical procedures is limited. A systematic review is performed from perspectives of the ECG database, preprocessing, DL methodology, evaluation paradigm, performance metric, and code availability to identify research trends, challenges, and opportunities for DL-based ECG arrhythmia classification. Specifically, 368 studies meeting the eligibility criteria are included. A total of 223 (61%) studies use MIT-BIH Arrhythmia Database to design DL models. A total of 138 (38%) studies considered removing noise or artifacts in ECG signals, and 102 (28%) studies performed data augmentation to extend the minority arrhythmia categories. Convolutional neural networks are the dominant models (58.7%, 216) used in the reviewed studies while growing studies have integrated multiple DL structures in recent years. A total of 319 (86.7%) and 38 (10.3%) studies explicitly mention their evaluation paradigms, i.e., intra- and inter-patient paradigms, respectively, where notable performance degradation is observed in the inter-patient paradigm. Compared to the overall accuracy, the average F1 score, sensitivity, and precision are significantly lower in the selected studies. To implement the DL-based ECG classification in real clinical scenarios, leveraging diverse ECG databases, designing advanced denoising and data augmentation techniques, integrating novel DL models, and deeper investigation in the inter-patient paradigm could be future research opportunities.

/pdf/deep-learning-based-ecg-arrhythmia-classification-a-2ys1u19a.pdf

Deep Learning-Based ECG Arrhythmia Classification: A Systematic Review

Electrocardiogram (ECG) signals carried important clinical information in the form of intervals and amplitude or morphology. Therefore, it is very important to identify these fiducial points effectively. The major limitations of the existing semi- and fully automatic ECG beat classification systems are extensive data requirements and low performance. This letter proposes a modified 1-D U-Net architecture to find the exact locations of the important waves in the ECG signal. The proposed network utilized 1-D convolutions instead of 2-D, which is different from the traditional U-Net. The encoder and decoder are the main components of the proposed 1-D U-Net, where the encoder is designed to extract high-level features of the input data. The encoder contains two downsampling stages, and the output of each stage in the encoder is sent to the corresponding decoder layer for further convolutional operations. The output of the decoder section concatenates with the respective encoder, which enables the network to perceive low-level and high-level features. The onset and offset of P, QRS, and T-waves in the ECG signal are identified using the proposed technique. The proposed method is validated on two publicly available databases, i.e., LUDB and MIT-BIH. The experimental results demonstrate that the performance of the proposed method is superior to the state-of-the-art techniques. 

A New Approach of Transparent and Explainable Artificial Intelligence Technique for Patient-Specific ECG Beat Classification

An end-to-end model for ECG signals classification based on residual attention network

Ventricular tachycardia (VT) and ventricular fibrillation (VF) are known ventricular cardiac arrhythmias (VCA) that promote fast defibrillation treatment for the survival of patients and are defined as shock-oriented signals, perhaps the most common source of sudden cardiac arrest (SCA). The majority of existing VCA classifiers confront a difficult challenge of accuracy rate, which has generated the issue of continuous detection and classification approaches. In light of this, the authors present a feature learning strategy that uses the improved variational mode decomposition technique to detect VCA on ECG signals. The following SCA consists of a deep convolutional neural network (deep CNN) as a feature extractor and bat-rider optimization algorithm (BROA) as an optimized classifier. The MIT-BIH arrhythmia database is used to examine the approaches, and the analysis depends on performance indicators such as accuracy, specificity, sensitivity, recall, and F1-score. 

Sudden Cardiac Arrest Detection by Feature Learning and Classification Using Deep Learning Architecture

OptRPC: A novel and optimized recurrence plot-based system for ECG beat classification

Early diagnosis of irregular cardiac activity through existing tools such as Electrocardiogram and greater understanding of the underlying processes is critical for saving lives. Cardiac activity originates from a deterministic dynamical system of heart with trajectories following a linear map. Irregular cardiac activity observed in arrhythmia patients adds nonlinearities to the evolution function of the dynamical system underneath. Therefore, it is of great importance to quantitatively measure this non-linearity as a biomarker for impending cardiac diseases in patients. In this work, we formulated a novel mechanism named Neural Temporal Perturbation Field where perceived nonlinearities are modeled through deep neural network with perturbated inputs. Here, we examined the nonlinear state space by modeling the volatility of outputs for slightly adjusted inputs. We discovered that volatility characteristics clearly define a decision threshold that may be employed as a biomarker in clinical practice by applying our technique to data on normal and abnormal heart activity. Our approach resulted in the greater understanding of nonlinearity and volatility of irregular cardiac activity and as a biomarker achieved comparatively better accuracy than the state-of-the-art models.

/pdf/inference-of-irregular-cardiac-activity-using-neural-1fjxk0lc.pdf

Mainul Islam Labib

Papers

OptRPC: A novel and optimized recurrence plot-based system for ECG beat classification

Inference of irregular cardiac activity using neural temporal perturbation fields