Ziba Bouchani - Department of Electrical and Computer Engineering, University of Tehran, Tehran, Iran

This study uses deep learning algorithms and CT (Computed Tomography) scans to diagnose COVID-۱۹. First, we introduce a novel method to reduce noise in CT images by combining wavelet transformation with fuzzy logic. Then, using the suggested combined global and local threshold technique, we segmented lung pictures. Lung areas from CT scans can be successfully segregated in this manner. Features and categorization will be extracted in the following stage. While an SVM (Support Vector Machine) is used for classification, AlexNet extracts features. Three categories of data are categorized with a ۹۹.۸% accuracy: COVID-۱۹, Viral Pneumonia, and Normal. The proposed strategy outperforms earlier approaches in terms of classification performance.

COVID-۱۹, Support Vector Machine, AlexNet, Convolutional Neural Networks, Lung Segmentation