Web server self-examination for detection of COVID-19 cases with fuzzy classifiers using chest X-ray image

ABSTRACT

The coronavirus has spread fast resulting in a worldwide pandemic. Early discovery of positive patients is critical in preventing the pandemic from spreading further, leading to the development of diagnostic technologies that provide rapid and reliable responses for COVID-19 detection. Previous research has shown that chest x-rays are an essential tool for the detection and diagnosis of sirivanoroC (COVID-19) patients. A radiological finding known as ground-glass opacity (GGO), which causes color and texture changes, was discovered in the lung of a person with COVID-19 as a consequence of x-ray tests. An automatic method to assist radiologists is required due to the carelessness of radiologists who work a long time and misdiagnosis resulting in the confusion of findings with different diseases, in this study, were described a new technique to help us with the early diagnosis of COVID-19 using x-rays that is based on fuzzy classification. The skewness, kurtosis, and average statistical features of x-rays of patients in two classes, COVID and Normal, are calculated in the suggested method, and the value ranges for both classes are identified. In the building of a fuzzy logic classifier, three statistical characteristics and value ranges are used as membership functions. The suggested solution, which uses a user-friendly interface, allows for quick and accurate COVID vs Normal (binary classification). Experiments show that our method has a lot of promise for radiologists to validate their initial screening and enhance early diagnosis, isolation, and therapy, which helps prevent infection and contain the pandemic. © 2023 Author(s).