COVID-19/pneumonia/normal classification from CXR images and U-net based COVID-19 segmentation with deep learning methods

The Covid-19 disease, named coronavirus, which emerged in China in December 2019, has become a pandemic in a short time all over the world. The fact that the Transcription Polymerase Chain Reaction (RT-PCR) test produces false negatives and the diagnosis time is long, has led to the search for new alternatives for the diagnosis of this virus, which can result in death, especially with the damage it causes to the lungs. Therefore, data obtained from CT or CXR imaging techniques of chest images have become suitable tools for diagnosis from chest images. Deep learning techniques studies have been proposed to provide diagnosis with these tools and to determine the infected area of Covid-19 and Pneumonia disease. In this thesis, a two-stage system is proposed as segmentation and classification. In the segmentation process, the infected regions segmented from the labeled data were determined. In the classifier stage, Covid-19/Pneumonia/Normal classification was performed using three different deep learning models named VGG16, ResNet50 and InceptionV3. As a result of the study, 95% segmentation accuracy was obtained. The classifier models obtained 99%, 90% and 98% accuracy, respectively.