Transfer Learning based on Covid-19 Classification using CT Scan Image

ABSTRACT

Initially, the coronavirus infection has been diagnosed by using the Chest CT scan and x-ray images of the patients. An accurate representation of the victim's respiratory system allows the medical practitioners to detect the covid-19 infection. The first step of the proposed approach is to preprocess the image in order to eliminate any undesirable noise that may be present in medical images. Following that, the intended features are retrieved from a processed image. Finally, Transfer Learning is used to categorize the data. The CT scan based representations are separated by using a U-net simulation, and the split representation is then used to train and analyze the data by using the v3 simulator, which helps to differentiate the coronavirus infection and pneumonia infection and securely protect the resulting documents. © 2022 IEEE.