Abstract
The novel COVID-19 pandemic has spread all over the world. Due to its easy transmission, it is crucial to develop techniques to accurately and efficiently identify the presence of COVID-19 and distinguish it from other forms of flu and pneumonia. Recent research has shown that the chest X-rays of patients suffering from COVID-19 depict specific radiography abnormalities. This study aims to construct a deep convolutional neural network (CNN) capable of performing feature extraction and binary classification of CT scans of COVID-19 patients from a publicly available dataset sourced from the University of California San Diego and Berkeley (UC San Diego & UC Berkeley). This work presents a 3-step technique to fine-tune pre-trained VGG19, Xception, and Inception V3 architectures to improve model performance and reduce training time. It was achieved by progressively re-sizing input images to 224x224x3 pixels and fine-tuning the network at each stage. Among three selected pre-trained models, the VGG 19 outperformed with 0.99, 0.88, 0.85, 0.86, 0.83, 0.85, 0.85 for Train accuracy, validation accuracy, test accuracy, precision, recall, F1 Score, the area under the curve values, respectively. Keywords: SARS-CoV2, Coronavirus, Deep Learning, Transfer Learning, Convolutional Neural Network DOI: 10.7176/JNSR/14-6-06 Publication date: April 30 th 2023
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.