Utilization of Transfer Learning Model in Detecting COVID-19 Cases From Chest X-Ray Images

Utilization of Transfer Learning Model in Detecting COVID-19 Cases From Chest X-Ray Images

Malathy Jawahar, L. Jani Anbarasi, Prassanna Jayachandran, Manikandan Ramachandran, Fadi Al-Turjman
Copyright: © 2022 |Pages: 11
DOI: 10.4018/IJEHMC.20220701.oa2
Article PDF Download
Open access articles are freely available for download

Abstract

Diagnosis of COVID-19 pneumonia using patients’ chest X-Ray images is new but yet important task in the field of medicine. Researchers from different parts of the globe have developed many deep learning models to classify COVID-19. The performance of feature extraction and classifier plays a vital role in the recognizing the different patterns in the image. The pivotal process is the extraction of optimum features from the chest X-Ray images. The main goal of this study is to design an efficient hybrid algorithm that integrates the robustness of MobileNet (using transfer learning approach) to extract features and Support Vector Machine (SVM) to classify COVID-19. Experiments were conducted to test the proposed algorithm and it was found to have a high classification accuracy of 95%.
Article Preview
Top

1. Introduction

Corona Virus Disease 2019 (COVID-19) is a new virus that was first discovered to be affecting humans in 2019 and it was found to be related to the same family of viruses as Severe Acute Respiratory Syndrome (SARS) (Lancet, 2020). It is a contiguous virus that have made its first appearance in Wuhan, China. Currently (as of 28 November 2020) 14,524,141 have died worldwide due to this virus. The rapid spread of this virus worldwide has resulted in a pandemic. India registered the first COVID-19 case on January 30 in Kerala and by February 03 two more cases were reported from the same city. In all three cases the infected individuals were students who have just returned back to India from Wuhan, China. The virus makes the immune system weaker which can eventually lead to death (Lancet, 2020; Razai et al., 2020). The virus is illustrated in Figure 1 and as it can be seen, the virus contain spikes on the crown of its outer surface which helps it in establishing a secured connection with human’s airway cells (Texas, n.d.). As of November 28, 2020; 62,165,882 people were infected worldwide and 42,951,570 have recovered (Jaiswal et al., 2019; Peng et al., 2020) in the world.

Figure 1.

Illustration of COVID-19 (Sharon et al., 2018)

IJEHMC.20220701.oa2.f01

COVID-19 can spread often by physical touch between two individuals. In general, it is possible for an infected induvial to infect other healthy individuals through breath, hand contact, touch, or mucosal contact (Peng et al., 2020). As the virus make its way to the lungs, infected individuals can suffer from pneumonia. Due to the extreme shortage of the expensive test kit, Real-time polymerase chain reaction (RT-PCR) (Xie et al., 2020) has further aggravated the situation. Hence, people with possible signs of pneumonia were prescribed a chest scan such as Computer Tomography (CT) scans and X-Rays to quickly diagnose and detect whether they are infected with COVID-19 or not. The presence of COVID-19 can be automatically detected with CT scans combined with deep learning techniques (Gozes et al., 2020; Li et al., 2020). In the recent years deep learning have demonstrated a promising result in various fields of research. For the medical field in particular, deep learning methodologies were found to deliver an improved accuracy on detecting different diseases using a data set of images such as images of chest X-Ray, retina image, and brain MRI (Mahmud et al., 2020; Mahmud et al., 2018). X-Ray machines are very useful as they offer a feasible and faster means of detecting disease through the process of scanning different human organs.

Complete Article List

Search this Journal:
Reset
Volume 15: 1 Issue (2024)
Volume 14: 1 Issue (2023)
Volume 13: 5 Issues (2022): 4 Released, 1 Forthcoming
Volume 12: 6 Issues (2021)
Volume 11: 4 Issues (2020)
Volume 10: 4 Issues (2019)
Volume 9: 4 Issues (2018)
Volume 8: 4 Issues (2017)
Volume 7: 4 Issues (2016)
Volume 6: 4 Issues (2015)
Volume 5: 4 Issues (2014)
Volume 4: 4 Issues (2013)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing