An Efficient Deep Learning Approach for Detection of COVID-19 from Chest CT Scan Images

Abstract

The virus is responsible for A person infected with the coronavirus may transmit the disease to others by the droplets released when they cough, sneeze, or breath. The two most common ways for a person to get the virus are via direct contact with an infected surface or by breathing in droplets that an infected person nearby releases into the air. When infected, the majority of people only have mild to moderate symptoms. The rapid onset of lung inflammation characterizes acute respiratory distress syndrome (ARSD), which is experienced by individuals with severe symptoms. Serious complications, including blood clots, multiple organ failures, and other life-threatening issues, may arise out of nowhere, and in extreme cases, the patient may not survive.Put another way, COVID-19 causes an unfavourable infection in our lungs, making it impossible for a person to breathe properly.. This paper's overarching goal is to examine and contrast CT scan pictures of individuals with coronavirus infections with those of people who may not have the virus or who may be suffering from other types of pneumonia. Also we have compared our present research VGG19 model with optimizer Adamax with our previous VGG19 model in this research work. The VGG19 model with optimizer Adamax give more accuracy and precision i.e. 85% and 89% respectively than our previous VGG19 model.