A deep learning computational approach for the classification of Covid-19 virus

Rajarajeswari, P, Santhi, K, Chattopadhyay, P and Beg, OA ORCID: https://orcid.org/0000-0001-5925-6711 2022, 'A deep learning computational approach for the classification of Covid-19 virus' , Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization .

[img] PDF (Deep learning algorithm for visualization of covid 19 scanning) - Accepted Version
Restricted to Repository staff only until 19 August 2023.

Download (766kB) | Request a copy

Abstract

Deep learning and transfer learning are being extensively adopted in biomedical, health and wellbeing related applications. The continuous Covid 2019 (COVID-19) pandemic brought about an extreme impact on the worldwide medical services framework, mainly as a result of its simple transmission and the all-encompassing time of the infection endurance on polluted surfaces. As per a worldwide agreement proclamation from the Fleischner Society, registered computer tomography (CT) is an applicable screening instrument owing to its higher efficiency in identifying early pneumonic changes since lung infection is a major manifestation of the covid 19 virus. Notwithstanding, doctors are still very involved battling COVID-19 in this period of overall emergency and new variants of the virus are emerging (delta, omicron) even after two years since the start of the pandemic. Hence, it is urgent to speed up the advancement of a man-made consciousness (AI) indicative device to help doctors. Regardless of colossal endeavors, it remains extremely difficult to create a powerful model to aid the exact measurement appraisal of COVID19 from the chest CT pictures. Due to the idea of obscured limits, regulated division techniques generally experience the ill effects of explanation predispositions. This image-based finding, it is envisaged will achieve significant improvements in more rapidly, effectively and accurately identifying Covid contamination in human beings. In this paper we have proposed CNN (convolutional neural network) based multi-picture growth procedure for recognizing COVID-19 in CT scans of Covid speculated patients. Multi-picture expansion utilizes irregularity data obtained from the shifted pictures for preparing the CNN model. We have proposed framework implements deep learning via multi-faceted CNN and with this methodology, the proposed

Item Type: Article
Additional Information: This is an Accepted Manuscript of an article published by Taylor & Francis in Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization on August 19th 2022, available at: http://www.tandfonline.com/10.1080/21681163.2022.2111722
Schools: Schools > School of Computing, Science and Engineering
Journal or Publication Title: Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization
Publisher: Taylor & Francis
ISSN: 2168-1163
Depositing User: OA Beg
Date Deposited: 22 Aug 2022 12:29
Last Modified: 30 Aug 2022 09:30
URI: http://usir.salford.ac.uk/id/eprint/64456

Actions (login required)

Edit record (repository staff only) Edit record (repository staff only)

Downloads

Downloads per month over past year