Online Automatic Diagnosis System of Cardiac Arrhythmias Based on MIT-BIH ECG Database

Yan, Wei; Zhang, Zhen

doi:10.1155/2021/1819112

Cited by 20 publications

(6 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the conventional methods, learning parameters during training the proposed techniques are able to cover multiple features with the confined nonlinear fitting and approximated capabilities in the facing of complex ECG waveforms. So, in the training of big data-driven context, the classification efficiency of conventional classifiers is not satisfactory [ 30 ].…”

Section: Introductionmentioning

confidence: 99%

An Effective and Lightweight Deep Electrocardiography Arrhythmia Recognition Model Using Novel Special and Native Structural Regularization Techniques on Cardiac Signal

Ullah

Heyat

AlSalman

et al. 2022

Journal of Healthcare Engineering

View full text Add to dashboard Cite

Recently, cardiac arrhythmia recognition from electrocardiography (ECG) with deep learning approaches is becoming popular in clinical diagnosis systems due to its good prognosis findings, where expert data preprocessing and feature engineering are not usually required. But a lightweight and effective deep model is highly demanded to face the challenges of deploying the model in real-life applications and diagnosis accurately. In this work, two effective and lightweight deep learning models named Deep-SR and Deep-NSR are proposed to recognize ECG beats, which are based on two-dimensional convolution neural networks (2D CNNs) while using different structural regularizations. First, 97720 ECG beats extracted from all records of a benchmark MIT-BIH arrhythmia dataset have been transformed into 2D RGB (red, green, and blue) images that act as the inputs to the proposed 2D CNN models. Then, the optimization of the proposed models is performed through the proper initialization of model layers, on-the-fly augmentation, regularization techniques, Adam optimizer, and weighted random sampler. Finally, the performance of the proposed models is evaluated by a stratified 5-fold cross-validation strategy along with callback features. The obtained overall accuracy of recognizing normal beat and three arrhythmias (V-ventricular ectopic, S-supraventricular ectopic, and F-fusion) based on the Association for the Advancement of Medical Instrumentation (AAMI) is 99.93%, and 99.96% for the proposed Deep-SR model and Deep-NSR model, which demonstrate that the effectiveness of the proposed models has surpassed the state-of-the-art models and also expresses the higher model generalization. The received results with model size suggest that the proposed CNN models especially Deep-NSR could be more useful in wearable devices such as medical vests, bracelets for long-term monitoring of cardiac conditions, and in telemedicine to accurate diagnose the arrhythmia from ECG automatically. As a result, medical costs of patients and work pressure on physicians in medicals and clinics would be reduced effectively.

show abstract

Section: Introductionmentioning

confidence: 99%

An Effective and Lightweight Deep Electrocardiography Arrhythmia Recognition Model Using Novel Special and Native Structural Regularization Techniques on Cardiac Signal

Ullah

Heyat

AlSalman

et al. 2022

Journal of Healthcare Engineering

View full text Add to dashboard Cite

show abstract

“…This section shows the classification results for controlling and recording items. For this purpose, a total of 582 recordings were selected from the previous database [33,34]. All signals were represented by selecting the number of diagnostic features defined in table 2.…”

Section: Resultsmentioning

confidence: 99%

Classification of electrocardiogram signals using deep learning based on genetic algorithm feature extraction

Khezripour¹,

Mozaffari²,

Reshadi³

et al. 2023

Biomed. Phys. Eng. Express

View full text Add to dashboard Cite

Arrhythmias using electrocardiogram (ECG) signal is important in medical and computer research due to the timely diagnosis of dangerous cardiac conditions. The current study used the ECG to classify cardiac signals into normal heartbeats, congestive heart failure, ventricular arrhythmias, atrial fibrillation arrhythmias, atrial flutter, malignant ventricular arrhythmias, and premature atrial fibrillation. A deep learning algorithm was used to identify and diagnose cardiac arrhythmias. We proposed a new ECG signal classification method to increase signal classification sensitivity. We smoothed the ECG signal with noise removal filters. A discrete wavelet transform based on an arrhythmic database was applied to extract ECG features. Feature vectors were obtained based on wavelet decomposition energy properties and calculated values of PQRS morphological features. We used the genetic algorithm to reduce the feature vector and determine the input layer weights of the artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS). Proposed methods for classifying ECG signals were in different classes of rhythm to diagnose heart rhythm diseases. Training data was with 80% of the data set and test data was with 20% for the whole data set. The learning accuracy for the results of training and test data in the ANN classifier was calculated as 99.9% and 88.92% and in ANFIS as 99.8% and 88.83% respectively. Based on these results, good accuracy was observed.

show abstract

“…Each portion of this wave contains information that is critical for the treating physician to make an accurate diagnosis. The faint and low frequency of the ECG signal, which spans from 0.5 Hz to 200 Hz, makes it very susceptible to interference from outside sources [9]. Abnormalities in the raw ECG signal include baseline wander noise, power line interference, and Electromyogram (EMG) noise.…”

Section: Introductionmentioning

confidence: 99%

Pre-processing of ECG signal based on multistage filters

Waheed

2023

JICS

View full text Add to dashboard Cite

In this paper, a new approach was adopted in order to remove the artifacts present in the raw ECG signal. To realizing this process a multiple types of filters for a variety of types of noises have been applied to obtain the lowest possible noise. Initially, the ECG signal was filtered from the electromyography signal, and then the filtered signal was passed again on another filter for the purpose of removing its drift from the baseline. The effectiveness of the suggested approach has been assessed in comparison to the conventional ECG signal. Results of this study showed significantly less signaling noise than was present in the conventional signal. Overall, the signal to noise ratio and the mean square error of the filtered ECG reached 15.72 and 0.006, respectively, in the filtered signal of one of the ECG recordings.

show abstract

Online Automatic Diagnosis System of Cardiac Arrhythmias Based on MIT-BIH ECG Database

Cited by 20 publications

References 20 publications

An Effective and Lightweight Deep Electrocardiography Arrhythmia Recognition Model Using Novel Special and Native Structural Regularization Techniques on Cardiac Signal

An Effective and Lightweight Deep Electrocardiography Arrhythmia Recognition Model Using Novel Special and Native Structural Regularization Techniques on Cardiac Signal

Classification of electrocardiogram signals using deep learning based on genetic algorithm feature extraction

Pre-processing of ECG signal based on multistage filters

Contact Info

Product

Resources

About