U-Net Based Covid-19 Infected Lesion Detection and Deep Learning Based Classification on CXR

The Covid-19 disease, which emerged in China in December 2019 and caused by the coronavirus virus, soon became a pandemic all over the world. The fact that the Transcription Polymerase Chain Reaction (RT-PCR) test produces false negatives in some studies 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, chest images have become suitable tools for diagnosis from chest images with data obtained from Computed Tomography or CXR imaging techniques. Deep learning studies have been proposed to provide diagnosis with these tools and to determine the infected region of Covid-19 and Pneumonia disease. In this paper, a two-stage system is proposed as segmentation and classification. In the segmentation process, 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. To the best of our knowledge, this is the first attempt to sequentially design classification and segmentation systems into a more precise diagnosis. As a result of the study, 95% segmentation accuracy was obtained. Classifier models achieved 99%, 90% and 98% accuracy, respectively. © 2022 IEEE.

A Method for Covid-19 Segmentation from X-Ray Images with U-Net

The Covid-19 virus, which emerged in China and affected the whole world, resulted in the death of many people in a short time and caused many socio-economic problems. This virus, which is mostly seen in patients with chronic diseases, has been seen worldwide in cases where it progressed rapidly and resulted in death in healthy individuals. Early diagnosis is one of the most important things to be done for this virus, which has such great effects. It is necessary to minimize the risk by treating the patient after being diagnosed and isolated early. The long time elapsed while providing diagnosis in current diagnostic methods potentially increases the course of the virus. For this reason, it has been deemed necessary to investigate some alternative ways for the diagnosis of Covid-19. In this sense, a study area has been created because radiological images have the defining characteristics of the virus. In this study, Covid-19, pneumonia and normal classification was made using X-Ray images. Then, we tried to determine the area affected by the Covid-19 virus using the U-Net system for image tissue classification. It is aimed to provide early detection and reduce workload with deep learning techniques to be used to solve these problems. © 2022 IEEE.

Detection of Covid-19 and Pneumonia from Colorized X-Ray Images by Deep Learning

The Coronavirus, which was first seen in Wuhan, China in December 2019, turned into an epidemic in a very short time. This virus, which has very serious effects especially in people with chronic diseases, causes global problems. The early diagnosis of Covid-19 and the isolation of the infected patient and then the treatment is very important. The inadequacy of diagnostic kits and the fact that radiological images contain the defining features of the virus have created a great field of study in this field. In addition to these, estimating the measures to be taken due to the increase in the number of cases is also important in terms of planning the processes related to many affected areas, especially the health field. Some models are used to take all these measures. In addition to these statistical models, deep learning-based artificial intelligence studies are also being developed. Particularly, the density in the field of health and the heavy workload of the employees have increased the tendency to develop artificial intelligence-based systems. In this context, the presence of unique symptoms related to Covid-19 in X-Ray images, which are frequently used in the diagnosis of many diseases in the field of health, has enabled the classification of these images with deep learning methods. Especially since Covid-19 and Pneumonia disease have similar patterns, it is necessary to classify these diseases by finding the uniqueness that even the human eye cannot see in the classification of these diseases. In this study, it is aimed to provide higher accuracy by colorizing X-ray images with deep learning methods and obtaining clearer images thanks to pre-trained networks. With this determination, the disease was classified as Covid-19/Pneumonia/Normal. In the study, the accuracy was provided with 98.78%.