ABSTRACT
"Over 186 million cases and 4.0 million deaths from the Corona virus were reported worldwide as of July 13, 2021. A multinational consensus from the Fleischner Society reported that Computerized Tomography (CT) can be utilized for the early classification of CT-based Covid-19. However, such diagnosis involves a significant amount of time by radiologists. Automated analysis to classify Covid-19 disease from lung CT will help save radiologists’ time and effort. In this paper, we propose ‘CoviNet Enhanced’, a hybrid approach with a deep three-dimensional convolutional neural network (3D-CNN) and support vector machines (SVMs) to diagnose Covid-19 in CT images, which is an improved version of our previous work ‘CoviNet’ based on only 3D-CNN. The experimental results show the proposed method is highly effective for Covid-19 detection.
