Prediction and Detection of Covid-19 through Deep Learning Technique

Coronavirus Disease (COVID19) is a rapidly spreading infectious disease which is producing a global healthcare crisis. Due to the present limitations of reverse transcription-polymerase chain reaction (RT-PCR)-based assays for identifying COVID19, numerous works have recently offered radiological imaging-based solutions. Computed Tomography (CT) imaging is useful for detecting COVID-19-related pulmonary symptoms in which segmentation of infection lesions from CT images is important for quantitative evaluation of disease progression and follow-up assessment. For this experimentation, 200 COVID-19 positive lung CT images and 200 COVID-19 negative lung CT images were collected. For validation, 110 images were obtained. The machine is made to learn the attributes of COVID-19 positive and COVID-19 negative images and is made to discriminate those using convolutional neural networks. As a result, when a new CT image is shown, the system will be able to anticipate whether the image is COVID-19 positive or negative.