Paper Contents
Abstract
The COVID-19 pandemic has drastically affected global health, leading to an urgent need for proper early detection and treatment tools. In this research work, image processing techniques are applied for the detection of COVID-19 from lung CT scans and the assessment of the degree of infection.The approach involves two phases, namely preprocessing of the CT images to enhance the important features that present COVID-19. The subsequently extracted features use deep learning models namely AlexNet, DenseNet-201, and ResNet-50. The extracted features are then evaluated using an Artificial Neural Network (ANN) to identify whether the patient is COVID-19 positive. If the infection has been determined, the next would be to determine the severity of the disease. The features of the image are combined with clinical data to categorize the severity of the infection into three levels-High, Moderate, and Low-using a Cubic Support Vector Machine SVM. The method was validated on several publicly available datasets, achieving accuracy of 92.0% in the detection of COVID-19 and a 90.0% accuracy in the categorization of the severity of the infection. These results suggest that this approach is a potential tool in helping healthcare providers more quickly and accurately diagnose COVID-19 and better manage patients.
Copyright
Copyright © 2024 Meghana Kalyanam. This is an open access article distributed under the Creative Commons Attribution License.
