What is the clinical evidence from this study?

Study design: Other. Population: Cardiac Arrhythmia (n=1000). Intervention: ArrhythmiaNet (Deep-Learning Framework) vs. Pre-trained D-CNNs and existing state-of-the-art models. Primary outcome: Classification accuracy.

March 10, 2022Open Access

A Novel Deep-Learning-Based Framework for the Classification of Cardiac Arrhythmia

Key Result

The proposed ArrhythmiaNet deep-learning framework achieved 99.84% accuracy, 100% sensitivity, and 99.6% specificity in classifying 17 classes of cardiac arrhythmia using 2D ECG signals.

Structured PICO

Population

1,000 ECG signal samples from the MIT-BIH dataset representing normal sinus rhythm and 16 classes of cardiac arrhythmia.

Intervention

ArrhythmiaNet, a novel deep-learning-based framework using continuous wavelet transform (CWT) to convert ECG into 2D signals, an attention-based deep convolutional neural network (D-CNN), and a clump of features (CoF) module with a Support Vector Machine (SVM) classifier.

Comparator

Pre-trained D-CNNs (AlexNet, ResNet-50, VGG-19, Inception v3, GoogLeNet, ShuffleNet, SqueezeNet, EfficientNetb0, Xception, DarkNet-53) and existing state-of-the-art machine learning techniques.

Outcome

Classification accuracy, sensitivity, specificity, F1-score, and classification error.

A novel deep-learning framework, ArrhythmiaNet, demonstrated 99.84% accuracy in classifying 17 classes of cardiac arrhythmias from 2D-transformed ECG signals.

Limitations

The proposed framework is dependent on the attention block, and the accuracy of the model decreases if the attention block is removed.
The performance of the model also varies by selecting different features.
The inter-patient paradigm is also a factor which affects the performance of the model.

Abstract

Cardiovascular diseases (CVDs) are the primary cause of death. Every year, many people die due to heart attacks. The electrocardiogram (ECG) signal plays a vital role in diagnosing CVDs. ECG signals provide us with information about the heartbeat. ECGs can detect cardiac arrhythmia. In this article, a novel deep-learning-based approach is proposed to classify ECG signals as normal and into sixteen arrhythmia classes. The ECG signal is preprocessed and converted into a 2D signal using continuous wavelet transform (CWT). The time-frequency domain representation of the CWT is given to the deep convolutional neural network (D-CNN) with an attention block to extract the spatial features vector (SFV). The attention block is proposed to capture global features. For dimensionality reduction in SFV, a novel clump of features (CoF) framework is proposed. The k-fold cross-validation is applied to obtain the reduced feature vector (RFV), and the RFV is given to the classifier to classify the arrhythmia class. The proposed framework achieves 99.84% accuracy with 100% sensitivity and 99.6% specificity. The proposed algorithm outperforms the state-of-the-art accuracy, F1-score, and sensitivity techniques.

Bookmark

View Full Paper