Arrhythmia Classification Using ECG Image Dataset Using Machine Learning Approach on DenseNet121 Model

In the fields of cardiology and biomedical engineering, categorization of arrhythmias is a subject of extensive investigation. The precise categorization of arrhythmias is crucial for diagnosis, therapy planning, and patient care. Arrhythmias are irregular cardiac rhythms that can have catastrophic...

Full description

Saved in:
Bibliographic Details
Published in2023 14th International Conference on Computing Communication and Networking Technologies (ICCCNT) pp. 1 - 4
Main Authors Gill, Kanwarpartap Singh, Anand, Vatsala, Gupta, Rupesh
Format Conference Proceeding
LanguageEnglish
Published IEEE 06.07.2023
Subjects
Arrhythmia
Classification
CNN
Computational modeling
Confusion matrix
Deep Learning
DenseNet121 Model
Electrocardiography
Machine learning
Medical treatment
Model Training
Telemedicine
Wearable computers
Online AccessGet full text

Cover

More Information
Summary:In the fields of cardiology and biomedical engineering, categorization of arrhythmias is a subject of extensive investigation. The precise categorization of arrhythmias is crucial for diagnosis, therapy planning, and patient care. Arrhythmias are irregular cardiac rhythms that can have catastrophic effects on a patient's health. Research on the categorization of arrhythmias frequently focuses on creating algorithms and models for precisely recognising and categorising various arrhythmia types using electrocardiogram (ECG) signals. This can aid in the identification of arrhythmias and assist physicians in deciding on the best course of therapy for patients. Research on the classification of arrhythmias frequently entails the creation and assessment of several feature extraction and selection strategies in order to pinpoint pertinent aspects from ECG signals that are most instructive for precise categorization. In order to create useful arrhythmia classification models, feature engineering is a crucial step. Research in this field aids in the discovery of novel features or feature combinations that might boost classification accuracy. The objective of this venture is to make an ECG classification based on exchange learning that can perceive arrhythmia. Our DenseNet121 model was shown to have good classification abilities for identifying arrhythmia, with an accuracy rate of more than 80%.
ISSN:2473-7674
DOI:10.1109/ICCCNT56998.2023.10308001