No Detail Found!

Use of deep neural networks with radiographic images for automated covid- 19 detection

Author(s)
JAVED M Department of Radiology, Children's Hospital & the Institute of Child Health Multan, Pakistan
ZAHRA M Department of Pediatric Radiology, Children's Hospital & the Institute of Child Health Multan, Pakistan
ALI M Department of Radiology, Children's Hospital & the Institute of Child Health Multan, Pakistan
Abstract
The prospective study was conducted in CH & ICH Multan from January 2021 to January 2022 to assess the significance of the proposed deep learning model that automatically uses X-rays to detect COVID-19. The development of the model proposed in the current study is based on the Darknet-19 model. When two classes are used, the proposed models detect COVID-19 infection. If three classes are used, the model classifies x-ray images as No finding, pneumonia, or COVID-19. First, the proposed model was used to classify X-ray images into Pneumonia, COVID-19, and No finding. Second, the model has been trained to detect two classes: No finding and COVID-19 categories. Our model achieved 87.06% and 97.88% accuracy for multiclass and binary tasks, respectively. Thus, it can be concluded that DarkCovidNet Deep Learning Model can be used for automated COVID-19 detection through X-ray images.
Publication Details
Page(s) 1-4
Published Journal: Biological and Clinical Sciences Research Journal, Volume: 4, Issue: 1, Year: 2023

Request the full text from PASTIC

Keywords
COVID-19, Deep neural network, Chest X-ray, automated detection
0

Citations

0

Downloads

37

Views