Detection and Analysis of Covid-19 from CT Scan images using Deep Learning based CNN Technique

In order to cure the COVID-19 pandemic, quick and precise COVID-19 identification from CT images that leverage deep identification of affected individuals is essential. This essay suggests a thorough method for learning methods. The processes covered by the suggested methodology are feature extraction, edge detection, preprocessing, segmentation, input image acquisition, and deep learning-based classification. First, as input, CT scans of possible instances are obtained. These pictures go through preprocessing to improve quality and reduce noise, and then they are segmented to separate lung areas. Then, lung borders are drawn using edge detection techniques for accurate feature extraction. Next, characteristics pertinent to COVID-19 infection are taken out of lung regions that have been segmented. These retrieved features are used to train deep learning models, like convolution neural networks (CNNs), which classify images into positive (COVID-19 infected) or negative (non-infected) categories. The effectiveness of the suggested method in correctly detecting COVID-19 instances from CT images is shown by the experimental findings. The deep learning model helps with effective disease diagnosis and management by achieving high accuracy, sensitivity, and specificity in differentiating between positive and negative cases. The suggested technique offers a quick and accurate way to screen and diagnose patients, making it a potentially useful tool for medical professionals fighting the COVID-19 pandemic.