Deep convolutional neural network application to classify the ECG arrhythmia

Abdalla, Fakheraldin Y. O.; Wu, Longwen; Ullah, Hikmat; Ren, Guoyu; Noor, Alam; Mkindu, Hassan; Zhao, Yaqin

doi:10.1007/s11760-020-01688-2

Cited by 25 publications

(15 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It allows this comparison to be fair in equal condition. Comparing to our proposed model with Abdalla [8] has marginally higher result than our approach. They utilized a CNN architecture with 11 layers construction.…”

Section: Methodsmentioning

confidence: 75%

“…They utilized two extensively common databases in their validation. Abdalla et al [8] adopted a convoluational neural network consisting of 11 layers for 10-class arrhythmia classification. They distributed these 11 layers into several parts to achieve higher accuracy.…”

Section: Related Workmentioning

confidence: 99%

“…Because of numerous contemporary medical applications where such problem may be mentioned, the relevance of ECG classification is currently quite high. However, the usage of heuristic hand-crafted or manufactured features with shallower feature learning frameworks is one of the primary drawbacks of these ML solutions [7, 8].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

HeartNet: Self Multi-Head Attention Mechanism via Convolutional Network with Adversarial Data Synthesis for ECG-based Arrhythmia Classification

Rafi

2021

Preprint

View full text Add to dashboard Cite

Cardiovascular disease is now one of the leading causes of morbidity and mortality in humans. Electrocardiogram (ECG) is a reliable tool for monitoring the health of the cardiovascular system. Currently, there has been a lot of focus on accurately categorizing heartbeats. There is a high demand on automatic ECG classification systems to assist medical professionals. In this paper we proposed a new deep learning method called HeartNet for developing an automatic ECG classifier. The proposed deep learning method is compressed by multi-head attention mechanism on top of CNN model. The main challenge of insufficient data label is solved by adversarial data synthesis adopting generative adversarial network (GAN) with generating additional training samples. It drastically improves the overall performance of the proposed method by 5-10% on each insufficient data label category. We evaluated our proposed method utilizing MIT-BIH dataset. Our proposed method has shown 99.67 ± 0.11 accuracy and 89.24 ± 1.71 MCC trained with adversarial data synthesized dataset. However, we have also utilized two individual datasets such as Atrial Fibrillation Detection Database and PTB Diagnostic Database to see the performance of our proposed model on ECG classification. The effectiveness and robustness of proposed method are validated by extensive experiments, comparison and analysis. Later on, we also highlighted some limitations of this work.

show abstract

Section: Methodsmentioning

confidence: 75%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

HeartNet: Self Multi-Head Attention Mechanism via Convolutional Network with Adversarial Data Synthesis for ECG-based Arrhythmia Classification

Rafi

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…In recent years, deep neural networks have been successfully introduced for the development of intelligent, rapid, and high-precision automatic arrhythmia classification that considers diverse detection parameters [1][2] [3]. For example, Fakheraldin et al [4] constructed a convolutional neural network (CNN) with 11 layers for classification; the classification accuracy of the network for 10 arrhythmia types from the MIT-BIH database was 99.84%, which is higher than the existing classification methods based on CNN. Huang et al [5] proposed an intelligent electrocardiogram (ECG) classifier based on a fast compressed residual convolution neural network (FCResNet).…”

Section: Introductionmentioning

confidence: 99%

Arrhythmia Classification Using Biased Dropout and Morphology-Rhythm Feature With Incremental Broad Learning

Zhang

Gao

et al. 2021

IEEE Access

View full text Add to dashboard Cite

In recent times, people have become increasingly health-conscious. To obtain timely and accurate information on the status of the heart, one of the most important organs of the human body, there is a growing demand among individuals and doctors for accurate and real-time automatic classification of arrhythmias. Consequently, this paper proposes a fast and accurate classification method for arrhythmias. In the proposed method, we build an incremental broad learning (IBL) classification model based on the biased dropout technique for arrhythmia-type recognition. In particular, we extract the morphological-rhythm features of the denoised signal as the input data of the IBL model in the electrocardiogram signal preprocessing. The IBL model enhances the classification effect of the node optimization model by using improved features. To the best of our knowledge, this study is the first application of the IBL model to the study of arrhythmia classification. The results of experiments conducted on the MIT-BIH database indicate that the proposed method is effective and achieves superior classification results. The average classification accuracy for six types of arrhythmias was 99%, and the training time required was only 2.7 s. In addition, based on the evaluation index recommended by the ANSI/AAMI EC57:2012 standard, our method is superior to existing methods on all indexes, except for the positive predictive rate of ventricular ectopic beats. Therefore, the proposed classification method outperforms state-of-the-art methods in terms of real-time performance and accuracy and provides a new approach for further improvements in arrhythmia classification. INDEX TERMSArrhythmia classification, biased dropout, broad learning, electrocardiogram, morphology-rhythm feature JIA LI received the Ph.D. degree in engineering from Jilin University, Changchun, China, in 2019. She received the M.Sc. degree from Kumamoto University, Kumamoto, Japan, in 2011, where she is currently a lecturer with the Department of Intelligent Manufacturing. Her research interests include digital image and biomedical signal processing, feature extracting, and classification in artificial intelligence.

show abstract

“…The many researchers used novel algorithms for detecting arrhythmia disease by means of ECG signals [8]. The time-domain features are extracted in a simple procedure from the segmented ECG signal to detect arrhythmia.…”

Section: Introductionmentioning

confidence: 99%

Multi-Class Arrhythmia Detection using a Hybrid Spatial-Temporal Feature Extraction Method and Stacked Auto Encoder

Sharmila¹

2021

IJIES

View full text Add to dashboard Cite

Cardiovascular disease generally refers to conditions that involve narrowed or blocked blood vessels which can lead to a heart attack, chest pain or stroke. The rate of mortality has been increased among cardiac patients and early detection of the heart disease is important to prevent death. The heart disease can generally be identified at an early stage by analysing irregular heart beat rhythms. The occurrence of abnormal heart beats is known as arrhythmia and the expert system is introduced for detection of arrhythmia. The Electro Cardio Gram (ECG) is considered to be a standard non-invasive diagnosis for Arrhythmia detection and classification of ECG signals was performed for the arrhythmia detection by the researchers in the existing methods. The existing Long Short Term Memory (LSTM)-Auto Encoder (AE) method was failed in selecting prior features to perform feature extraction and also gave rise to overfitting problems. The high level features were need to tackle the issue occurred during data imbalance which gave rise to misclassification of ECG heartbeats. In order to overcome aforementioned issue, the proposed hybrid spatial temporal feature extraction technique extracts salient features and eliminates the exhausted handcrafted features. Initially the ECG signals are acquired from the Massachusetts Institute of Technology-Beth Israel Hospital (MIT-BIH) database. The ECG signals were undergone for the pre-processing stage using butter worth filter. The segmentation of the QRS complex is an important task as the QRS spike in signal determines the abnormality in heart beat using Teager Energy Operator (TEO). The spatial features that provide locations of various natural or artificial boundaries or shapes visualize spatial data of the ECG signals. The temporal features are the time domain features that include Energy of Signal, Zero Crossing Rate, and Maximum Amplitude parameters. These parameters from both spatial and temporal feature forms a hybrid spatial temporal features that are extracted from the QRS complex and are fed to the layers of Deep Neural Network (DNN) based Stacked Auto Encoder (SAE) The hybrid spatial-temporal features determines the type of heart rhythm into 5 prominent classes of Arrhythmia.

show abstract

Deep convolutional neural network application to classify the ECG arrhythmia

Cited by 25 publications

References 26 publications

HeartNet: Self Multi-Head Attention Mechanism via Convolutional Network with Adversarial Data Synthesis for ECG-based Arrhythmia Classification

HeartNet: Self Multi-Head Attention Mechanism via Convolutional Network with Adversarial Data Synthesis for ECG-based Arrhythmia Classification

Arrhythmia Classification Using Biased Dropout and Morphology-Rhythm Feature With Incremental Broad Learning

Multi-Class Arrhythmia Detection using a Hybrid Spatial-Temporal Feature Extraction Method and Stacked Auto Encoder

Contact Info

Product

Resources

About