COVID-19 Detection Using Deep Learning: A Comparative Study of Segmentation Algorithms

ABSTRACT

Lung abnormality is a prevalent condition that affects people of all ages, and it can be caused by a variety of factors. The lung illness caused by SARS-CoV-2 has recently spread across the globe, and the World Health Organization (WHO) has labelled it a pandemic disease owing to its quickness. Covid-19 mainly attacks the lungs of those infected, resulting in mortality from ARDS and pneumonia in extreme instances. Internal body organ disorders are thought to be more acute, making diagnosis more complex and time-consuming. The source of any illness, location and severity are determined by a pulmonologist basing upon a good number of tests taken in the laboratories or even outside these after the hospitalization of a patient. In between a lot of time is taken to carry out these tests and prediction of COVID 19 is done. The purpose of this work is to propose a model based on CNN and finding out the best fit segmentation algorithm to apply to the chest X-ray scans in order to predict the test result. Most importantly the result is instantaneous. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.