Novel Architecture for Image Classification Based on Rough Set

Novel Architecture for Image Classification Based on Rough Set

S. Nivetha, H. Hannah Inbarani
DOI: 10.4018/IJSSMET.323452
Article PDF Download
Open access articles are freely available for download

Abstract

The Computed Tomography (CT) scan images classification problem is one of the most challenging problems in recent years. Different medical treatments have been developed based on the correctness of CT scan images classification. In this work, a novel deep learning architecture is proposed to correctly diagnose COVID-19 patients using CT scan images. In fact, a new classifier based on rough set theory is suggested. Extensive experiments showed that the novel deep learning architecture provides a significant improvement over well-known classifier. The new classifier produces 95% efficiency and a very low error rate on different metrics. The suggested deep learning architecture coupled with novel tolerance outperforms the other standard classification approaches for the detection of COVID-19 using CT-Scan images.
Article Preview
Top

1. Introduction

Many pandemic viruses, such as smallpox (Henderson, 2009) and Spanish influenza (Spreeuwenberg et al., 2018) have killed millions of people in recent years. In the years 2003, 2009, and 2012, various pandemics were observed, including H1N1 influenza (Pandemic H1N1 2009), Severe Acute Respiratory Syndrome Corona Virus (SARS-CoV) (Pandemic SARS., 2003), and the Middle East Respiratory Syndrome Corona Virus (MERS-CoV) (Pandemic MER’s-cov., 2012). SARS and MERS have alike symptoms, but SARS is more virulent, and MERS is more contagious. The disease was initially unknown, but experts identified its signs as being similar to influenza and coronavirus diseases (Chen et al., 2020; Huang et al., 2020 ; Lima, 2020; Struyf et al., 2020). As of November 2021, there were more than 253,684,701 confirmed cases of the disease, 5,110,870 are death cases, and 229,391,450 recovered cases are addressed all around the world. First, the people are infected with the virus and act as carriers, but they may not show any symptoms. These people are more likely to spread the virus since they may be unaware of its presence.

Individuals with a minor fever, cough, headache, or probable conjunctivitis fall into the second category. An upper respiratory tract infection is the main indicator in the second category. In the third category of people, the symptoms are similar to those in the second group, but they are more severe and may necessitate hospitalization. Treatment as soon as possible can help reduce symptoms and prevent death. Acute cases of COVID 19 might reason Acute Respiratory Distress Syndrome (ARDS) and Pneumonia in the fourth category. It is deadly at this stage. This contamination is highly infectious and spreads even more rapidly through respiratory droplet infection compared to other kinds of flu. Though patches with COVID-19 pneumonia with a specific pattern may be seen on CT scans by naked human eyes, it is simple to overlook those minor and minimally diseased areas, especially in the initial stages. Because of this, radiologists need to be adequately trained to make an early and accurate diagnosis, which is crucial for prompt treatment as well as for population screening and response. Initial COVID-19 detection is essential for both patient care and epidemiology because it enables individuals to be isolated and the epidemic to be contained (World Health Organization, 2020; Wang et al., 2020; Azar & Hassanien, 2022). However, due to the time-consuming and difficult aspect of professional training, there is a dearth of qualified radiologists, making correct diagnoses all the more difficult given the new increase in cases. An automated and reliable approach is urgently needed to increase the consistency and ease of COVID-19 CT based diagnosis. The United States, the United Kingdom, and a number of other nations have adopted and utilized vaccines made by Pfizer (USA), AstraZeneca (UK), and Moderna (USA). Based on the findings of clinical trials, the three effective vaccinations are thought to have attained the target of 50% effectiveness and are safe to use without any noticeable adverse effects. (Ledford et al., 2020; Jafari et al., 2022). Many countries, including some Latino countries, such as Columbia, Uruguay, Chile, and Brazil along with other countries such as Laos, Turkey and Indonesia have recently approved the vaccine produced by Sinovac Life Sciences in China (Kim et al., 2021)

Complete Article List

Search this Journal:
Reset
Volume 15: 1 Issue (2024)
Volume 14: 1 Issue (2023)
Volume 13: 6 Issues (2022): 2 Released, 4 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