Revolutionizing Covid-19 Pneumonia Detection with Convolutional Neural Networks

Since the currently available vaccinations do not prevent the disease but only lessen the intensity of the symptoms, it has become clear that humanity must learn to coexist with and get used to Covid-19. It takes a long time and is highly expensive to manually classify and diagnose COVID-19 pneumonia. Conversely though, automated diagnosis would provide instantaneous diagnosis without human involvement, results in lower expenses. As a result, the goal of thisresearch is to provide a unique, optimize Deep Learning method for automatically diagnosing and classifying covid-19 pneumonia from X-ray pictures. This paper used accessible to the public images of chest x-rays from Kaggle for this reason in an effort to provide researchers with The dataset, united covid-19 entities, was created over the course of three stages. The dataset includes 21000 chest X-ray pictures from the front to the back and back to the front, categorized as normal, covid-19, Lung Opacity and Viral Pneumonia. In order to expand the size of the dataset and improve the trust worthiness of the outcomes by avoiding overfitting, data augmentation are also used. Chest X-ray pictures are processed in three stages as part of an optimal DL technique. The first stage involves Image Enhancement, the second involves Data Augmentation, and the third and final stage involves feeding the findings to the Transfer Learning algorithms classify and diagnose the photos.