A Branched Deep Neural Network for End-to-end Classification from ECGs with Varying Dimensions

Duan, Han; Fan, Junchao; Xiao, Bin; Bi, Xiuli; Zhang, Junhui; Ma, Xu

doi:10.23919/cinc53138.2021.9662727

Cited by 1 publication

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared to SENet-34 and ResNet-34, the proposed classification network achieved an F1 score of 0.028, higher than SENet-34 and 0.125 higher than ResNet-34. Moreover, the proposed classification network exhibits higher F1 scores of 0.07, 0.166, 0.152, and 0.098 compared to other CNN-based networks [19], [20], [21], and [23], respectively.…”

Section: Implementation Results Using a Large-scale 12-lead Ecg Datab...mentioning

confidence: 88%

“…However, the proposed classification network outperforms other CNN-based classification networks because it performs best in the most important evaluation metric, the F1 score. As a result, the proposed network shows superior performance in terms of precision compared to SENet-34, ResNet-34, [19], [21], and [23]. However, the precision lags slightly [20] by 0.020.…”

Section: E Implementation Results Using Cpsc 2018 Datasetmentioning

confidence: 90%

“…Specifically, comparing the AUROCs of SENet-34 and ResNet-34 improved by 0.003 and 0.087, respectively. When compared with the other four cardiac arrhythmia classification networks [19], [20], [21], and [23], it can be seen that the AUROC is improved by 0.003, 0.074, 0.070, and 0.034, respectively. In addition, improvements of 0.001, 0.099, 0.022, 0.150, 0.074, and 0.046 were observed when comparing the six different classification networks in accuracy, respectively.…”

Section: Implementation Results Using a Large-scale 12-lead Ecg Datab...mentioning

confidence: 97%

“…Their proposed DNN could classify nine rhythm classes using the CPSC 2018 dataset and achieved superior performance than four ML methods on average F1 score. Duan et al [21] proposed an end-to-end DNN structure based on SENet [31] to classify This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and content may change prior to final publication.…”

Section: Convolutional Neural Network-based Methodsmentioning

confidence: 99%

“…denotes concatenated sum. As a result, the proposed automated multi-label cardiac arrhythmia classification network could achieve higher performance compared to other existing networks [19]- [21], and [23]. Detailed explanations regarding the experimental results are provided in Section IV.…”

Section: A the Proposed Classification Network Architecture To Classi...mentioning

confidence: 99%

See 4 more Smart Citations

An Automated Cardiac Arrhythmia Classification Network for 45 Arrhythmia Classes Using 12-Lead Electrocardiogram

Kim,

Sunwoo

2024

IEEE Access

View full text Add to dashboard Cite

Electrocardiogram is a non-invasive, inexpensive, and widely used diagnostic tool for arrhythmia diagnosis in clinics. Deep learning techniques have shown great promise in electrocardiogram signal analysis, enabling automatic and accurate detection of various cardiac arrhythmia. This paper proposes an automated multi-label cardiac arrhythmia classification network based on a convolutional neural network. The network aims to detect and classify 45 different cardiac arrhythmia classes using 12-lead electrocardiogram data. Unlike previous studies, our approach incorporates both the residual structure and channel attention mechanism. Thus, we developed two key schemes to improve classification performance: the Global Channel Attention Block and the Short Residual Block. The Global Channel Attention Block incorporates dilated convolutions to preserve overall features. It focuses on the important characteristics of each arrhythmia class from the original electrocardiogram data during the training process. The Short Residual Block employs a residual structure to enhance classification accuracy. The network's performance is evaluated using a large-scale 12-lead electrocardiogram database for arrhythmia study on PhysioNet and the 2018 China Physiological Signal Challenge dataset. In particular, the proposed classification network shows the highest scores in average precision, recall, F1 score, area under the receiver operating characteristic, and accuracy compared to existing convolutional neural network-based arrhythmia classification networks in a large-scale 12-lead electrocardiogram database for arrhythmia study on PhysioNet.

show abstract

Section: Implementation Results Using a Large-scale 12-lead Ecg Datab...mentioning

confidence: 88%

Section: E Implementation Results Using Cpsc 2018 Datasetmentioning

confidence: 90%

Section: Implementation Results Using a Large-scale 12-lead Ecg Datab...mentioning

confidence: 97%

Section: Convolutional Neural Network-based Methodsmentioning

confidence: 99%

Section: A the Proposed Classification Network Architecture To Classi...mentioning

confidence: 99%

See 3 more Smart Citations

An Automated Cardiac Arrhythmia Classification Network for 45 Arrhythmia Classes Using 12-Lead Electrocardiogram

Kim,

Sunwoo

2024

IEEE Access

View full text Add to dashboard Cite

show abstract

A Branched Deep Neural Network for End-to-end Classification from ECGs with Varying Dimensions

Cited by 1 publication

References 9 publications

An Automated Cardiac Arrhythmia Classification Network for 45 Arrhythmia Classes Using 12-Lead Electrocardiogram

An Automated Cardiac Arrhythmia Classification Network for 45 Arrhythmia Classes Using 12-Lead Electrocardiogram

Contact Info

Product

Resources

About