Classification of ECG signals using one-dimensional hybridconvolutional neural networks and long short-term memoryarchitectures

The demand for effective monitoring of sub-health problems is rapidly increasing. Heart diseases are the leading causes of death today. For this reason, studies on the classification of heartbeats have recently attracted great attention. Electrocardiography is an easy and effective way to identify, detect and diagnose cardiac arrhythmia. It is difficult to obtain information about heart diseases from electrocardiographic (ECG) signals. An accurate model to detect abnormal ECG signals enables early diagnosis of disease and the right treatment for patients. In this study, we will use a 1D deep learning model combining convolutional neural network (CNN) and long short-term memory (LSTM) methods for accurate, fast and automatic beat ECG classification. Our goal is to determine the accuracy of our model by applying this model to the relevant ECG datasets and obtain a classification with high accuracy. Keywords: Deep Learning, Electrocardiography, Arrhythmia Detection, Convolutional Neural Networks (CNN), 1D Convolutional Neural Networks (1D CNN), Long Short Term Memory (LSTM)