Pneumonia Severity Diagnosis: A Deep Learning Perspective

• Author(s): Olatunde Olukemi Victoria
• Paper ID: 1705444
• Page: 414-418
• Published Date: 25-01-2024
• Published In: Iconic Research And Engineering Journals
• Publisher: IRE Journals
• e-ISSN: 2456-8880
• Volume/Issue: Volume 7 Issue 7 January-2024

Abstract

Pneumonia has been responsible for a large number of under-five children deaths worldwide. The most common traditional method of diagnosing pneumonia is through X-ray Images, however, Pneumonia shares common features with other respiratory diseases, such as lung cancer and bronchitis, which make it difficult to distinguish pneumonia from them. Also, there is significant flexibility in the way chest X-ray (CXR) images are acquired and processed, which can greatly impact the quality and consistency of the images. This can cause difficulty in developing powerful algorithms that can accurately identify pneumonia in all types of images. Deep Learning (DL) models with Convolutional Neural Networks (CNN) as the most commonly employed model have significantly contributed to the diagnosis and classification of chest X-ray images into healthy and pneumonia classes. However, there is a need to go beyond binary classification by knowing the severity levels of those diagnosed with pneumonia for faster treatment. This paper proposes a pneumonia severity diagnosis model using a deep learning approach that will assist medical practitioners prioritise pneumonia patients’ treatment.

Keywords

CNN, CXR, DL, Pneumonia, Severity.

Citations

IRE Journals:
Olatunde Olukemi Victoria "Pneumonia Severity Diagnosis: A Deep Learning Perspective" Iconic Research And Engineering Journals Volume 7 Issue 7 2024 Page 414-418

IEEE:
Olatunde Olukemi Victoria "Pneumonia Severity Diagnosis: A Deep Learning Perspective" Iconic Research And Engineering Journals, 7(7)