COVID-19 diagnosis in CXR Images using Deep Transfer Learning Approach

The COVID-19 virus is a novel pathogen that has genetic similarities with SARS and several cold viruses. The utilization of traditional test procedures including polymerase chain reactions, serology tests, and antigen assays is common. The patient may not receive the test results for several days or it may take just several hours. Hence, the implementation of an autonomous diagnosis system as a quick novel diagnostic system is required to prevent the spread of COVID-19 between the population. Although a number of researchers had excellent success in the detection of COVID-19, the majority of them had lower accuracy and overfitting problems that make advance screening of this is challenging. The best method for more accurately resolving this issue is transfer learning. In this study, four convnets supported by pre-trained neural networks (ResNet50, DenseNet101, Inception-V3 and CapsNet) are presented for identifying patients with pneumonia due to bacterial or coronavirus or any other virus or normal using Chest X-Rays (CXR). The proposed study implements multi class classifications using cross-validation. When results are taken into account, the pre-trained ResNet50 model offers the best classification efficiency (97.77% accuracy, 100% sensitivity, 93.33% specificity, and 98.00% F1-score) over 6259 images of the other four models that were utilized. © 2022 IEEE.

Classification of Lung diseases using Convolutional Neural Network algorithms

Since the publication of the book Covid-19, several investigations of varying kinds have been carried out all across the globe to see how well it predicted future events. The early lung illness known as pneumonia is intimately linked to the virus known as Covid-19, which causes severe inflammation of the chest (pneumonic condition). It is difficult for doctors and other medical professionals to differentiate Covid-19 from other lung diseases including pneumonia. As a consequence of this, we need an independent diagnostic platform that is able to provide clinical results in a timely and efficient manner. Chest X-ray screening is the method that provides the most reliable diagnosis of lung disease. The purpose of this investigation was to offer a condensed CNN (RMNet) model for COVID-19 classification. When compared to prior models, the solution that has been developed requires less memory and requires fewer processing resources. When it comes to COVID-19 classification, the performance of the recommended RMNet model ensemble that makes use of ResNet18, Inceptionv3, and MobileNetV2 is superior to that of previously cutting-edge methodologies. Additionally, the ensemble model makes less of a demand on available memory and is straightforward to incorporate into the backend of a smart device. Lung cancer is produced by the unrestrained growth of aberrant cells. This proliferation may begin in either of the lungs, but it most often originates in the cells that border the airways. Lung cancer can be prevented by maintaining a healthy immune system. These abnormal cells do not develop into lung tissue that is healthy;instead, they rapidly multiply and produce tumours. This is because of how they behave. The process by which cancer cells spread from the primary site of the disease to other parts of the body is referred to as "metastasis."Once the disease has spread to other areas of the body, it is much more challenging to treat it in an appropriate manner. Primary or secondary lung cancer is a classification that may be used to this disease. Primary lung cancer starts in the lungs, but secondary lung cancer starts elsewhere in the body, metastasizes, and then spreads to the lungs. Both types of lung cancer may be fatal. Because medical professionals consider them to be separate manifestations of the illness, they are not treated the same way because of this belief. The information offered by symptoms, in addition to the findings of a number of other tests, is taken into consideration by medical professionals when making a diagnosis of lung cancer. Imaging techniques such as chest X-rays, bronchoscopies, CT scans, MRI scans, and PET scans are examples of what are known as conventional imaging methods. In addition, the doctor will do a physical examination on the patient, as well as an inspection of the chest, and a test to determine whether or not there is blood in the sputum. The goal of each of these procedures is to zero in on the specific location of the tumour and determine whether other organs in the body may be at risk due to the presence of the malignant growth. © 2022 IEEE.

Deep Learning and Classification Algorithms for COVID-19 Detection

The imaging modalities of chest X-rays and computed tomography (CT) are commonly utilized to quickly and accurately diagnose COVID-19. Due to time and human error, it is exceedingly difficult to manually identify the infection using radio imaging. COVID-19 identification is being mechanized and improved with the use of artificial intelligence (AI) tools that have already showed promise. This study employs the following methodology: The chest footage was pre-processed by setting equalizing the histogram, sharpening it, and so on. The transformed chest images are then retrieved through shallow and high-level feature mapping over the backbone network. To further improve the classification performance of the convolutional neural network, the model uses self-attained mechanism through feature maps. Numerous simulations show that CT image classification and augmentation may be accomplished with higher efficiency and flexibility using the Inception-Resnet convolutional neural network than with traditional segmentation methods. The experiment illustrates the association between model accuracy, model loss, and epoch. Inception-statistical Resnet's measurement results are 98%, 91%, 91%. © 2022, International Journal of Advanced Computer Science and Applications. All Rights Reserved.