An Embedded System Using Convolutional Neural Network Model for Online and Real-Time ECG Signal Classification and Prediction

Abstract

This paper presents an automatic ECG signal classification system that applied the Deep Learning (DL) model to classify four types of ECG signals. In the first part of our work, we present the model development. Four different classes of ECG signals from the PhysioNet open-source database were selected and used. This preliminary study used a Deep Learning (DL) technique namely Convolutional Neural Network (CNN) to classify and predict the ECG signals from four different classes: normal, sudden death, arrhythmia, and supraventricular arrhythmia. The classification and prediction process includes pulse extraction, image reshaping, training dataset, and testing process. In general, the training accuracy achieved up to 95% after 100 epochs. However, the prediction of each ECG single type shows a differentiation. Among the four classes, the results show that the predictions for sudden death ECG waveforms are the highest, i.e., 80 out of 80 samples are correct (100% accuracy). In contrast, the lowest is the prediction for normal sinus ECG waveforms, i.e., 74 out of 80 samples are correct (92.5% accuracy). This is due to the image features of normal sinus ECG waveforms being almost similar to the image features of supraventricular arrhythmia ECG waveforms. However, the model has been tuned to achieve an optimal prediction. In the second part, we presented the hardware implementation with the predictive model embedded in an NVIDIA Jetson Nanoprocessor for the online and real-time classification of ECG waveforms.

Comparison of Three Prototypes of PPG Sensors for Continual Real-Time Measurement in Weak Magnetic Field

“…On the other hand, the first- and the second-order derivatives of this PPG signal are more informative due to more pronounced local extremes. The PPG waveform for the purpose of classification or prediction processes could also be used in the form of an image in a similar way as this approach is applied in the case of real-time electrocardiogram processing [ 16 ].…”

ASCNet-ECG: Deep Autoencoder based Attention aware Skip Connection network for ECG filtering

“…Due to their wide use in the biomedical field, several researchers have carried out to analyse heart activities effectively. ECG signals are widely adopted in various applications such as Caesarendra et al [10] developed CNN based model to detect normal heart health. Supraventricular arrhythmia using ECG signal, Parmar et al [11] implemented the Fourier decomposition method and modulated filter bank to detect hypertension, Ayashm et al [12] analysed ECG signal [8] to detect sleep Apnea, Ramkumar et al [13] used ECG processing for arrhythmia classification.…”

A Robust Framework Combining Image Processing and Deep Learning Hybrid Model to Classify Cardiovascular Diseases Using a Limited Number of Paper-Based Complex ECG Images

“…Several researchers have proposed various deep learning methods to classify cardiac diseases using ECG images or ECG signals. We reviewed some studies [ 15 , 16 , 17 , 18 ] that applied different deep learning techniques to paper-based ECG images and some studies [ 19 , 20 , 21 , 22 , 23 , 24 ] that used digital ECG signals to classify cardiac diseases. Mehmet et al [ 15 ] used the cardiac and COVID-19 paper-based ECG image dataset to classify COVID-19 ECGs.…”

“…Several other studies used digital ECG signals for disease classification. Wahyu et al [ 19 ] proposed a method to detect and predict diseases for real-time ECG signals using the PhysioNet open-source database. The authors first used a CNN model to classify and predict four different classes: normal, sudden death, arrhythmia, and supraventricular arrhythmia.…”