A Hybrid Deep Learning Approach for ECG-Based Arrhythmia Classification

Abstract

Arrhythmias are defined as irregularities in the heartbeat rhythm, which may infrequently occur in a human’s life. These arrhythmias may cause potentially fatal complications, which may lead to an immediate risk of life. Thus, the detection and classification of arrhythmias is a pertinent issue for cardiac diagnosis. (1) Background: To capture these sporadic events, an electrocardiogram (ECG), a register containing the heart’s electrical function, is considered the gold standard. However, since ECG carries a vast amount of information, it becomes very complex and challenging to extract the relevant information from visual analysis. As a result, designing an efficient (automated) system to analyse the enormous quantity of data possessed by ECG is critical. (2) Method: This paper proposes a hybrid deep learning-based approach to automate the detection and classification process. This paper makes two-fold contributions. First, 1D ECG signals are translated into 2D Scalogram images to automate the noise filtering and feature extraction. Then, based on experimental evidence, by combining two learning models, namely 2D convolutional neural network (CNN) and the Long Short-Term Memory (LSTM) network, a hybrid model called 2D-CNN-LSTM is proposed. (3) Result: To evaluate the efficacy of the proposed 2D-CNN-LSTM approach, we conducted a rigorous experimental study using the widely adopted MIT–BIH arrhythmia database. The obtained results show that the proposed approach provides ≈98.7%, 99%, and 99% accuracy for Cardiac Arrhythmias (ARR), Congestive Heart Failure (CHF), and Normal Sinus Rhythm (NSR), respectively. Moreover, it provides an average sensitivity of the proposed model of 98.33% and a specificity value of 98.35%, for all three arrhythmias. (4) Conclusions: For the classification of arrhythmias, a robust approach has been introduced where 2D scalogram images of ECG signals are trained over the CNN-LSTM model. The results obtained are better as compared to the other existing techniques and will greatly reduce the amount of intervention required by doctors. For future work, the proposed method can be applied over some live ECG signals and Bi-LSTM can be applied instead of LSTM.

Bimodal CNN for cardiovascular disease classification by co-training ECG grayscale images and scalograms

“…In this way, Madan et al introduced a hybrid model called 2D-CNN-LSTM, which transformed the ECG signals into scalograms and then combined two learning models, CNN and LSTM. The results obtained were better than other conventional techniques 30 . In a similar way, Jeong & Lim converted 12-channel ECG recordings into time–frequency feature map through short-time Fourier transform (STFT) 17 .…”

“…Therefore, numerous preprocessing processes such as filtering and noise removal are required to ensure data integrity and improve model accuracy. To mitigate this problem, one-dimensional ECG recordings can be converted into scalograms or spectrograms which automate the noise filtering and feature extraction 30 . In this way, Madan et al introduced a hybrid model called 2D-CNN-LSTM, which transformed the ECG signals into scalograms and then combined two learning models, CNN and LSTM.…”

Multi-Label Attribute Selection of Arrhythmia for Electrocardiogram Signals with Fusion Learning

“…Although most neural networks can reduce the preprocessing, the preprocessing still significantly improves the final ECG recognition and classification capabilities [ 29 , 30 ]. The mixture of multiple noises will affect the recognition and the feature is difficult to extract [ 31 ].…”

ECG Paper Digitization and R Peaks Detection Using FFT

“…Madan et al [16] offered a model based on a hybrid deep learning called 2-D-CNN-LSTM for the disclosure and rating process. The methodology of this paper includes two stages: automated noise reduction and characteristic extraction.…”