| ECG Signal Classification Based on Deep Learning: A Survey |
| Paper ID : 1099-ICEEM2023 |
| Authors: |
|
Doaa Kattab1, Sayed M. El-Rabie2, Fathi E. Abd El-Samie3, Heba M. Emara *4 1Menoufia Unversity, Faculty of electronic Engnieering, Egypt. 2Faculty of Electronic Engineering, Menoufia University, Egypt. 3Faculty of Electronic Engineering, Menoufia University 4Faculty of Electronic Engineering, Menoufia University, Menouf, Egypt. |
| Abstract: |
| Electrocardiogram (ECG) classification plays a vital role in modern healthcare, enabling early detection and diagnosis of cardiac arrhythmias, which are critical in preventing lifethreatening cardiac events. The increasing interest in leveraging machine learning and artificial intelligence in medical applications has led to a plethora of ECG classification techniques. This paper presents a comprehensive survey of ECG classification methods and their efficacy in arrhythmia detection. The objective of the survey is to explore the diverse array of approaches employed in ECG classification, ranging from signal preprocessing and feature extraction to various machine learning algorithms. By analyzing and comparing the performance of these methods, valuable insights into the strengths, limitations, and future prospects of ECG classification for improved medical diagnosis are provided. The survey encompasses an extensive review of relevant literature, including recent studies, to present a comprehensive assessment of the state-of-the-art in ECG classification. The findings of this survey have the potential to guide researchers, clinicians, and developers in selecting the most appropriate techniques and facilitating advancements in cardiac healthcare through automated arrhythmia detection systems. |
| Keywords: |
| Electrocardiogram (ECG), ECG classification, Arrhythmia detection Machine learning, Deep learning |
| Status : Paper Accepted |
