A FRAMEWORK OF CONVONET FOR DETECTION OF COVID-19 IN CHEST X-RAY IMAGE

Abstract

COVID-19 cases are increasing worldwide day by day leads to a huge load on health amenities. Due to the limited accessibility of reverse transcription-polymerase chain reaction (RT-PCR) kits, every patient with respiratory illness could not get tested by it. The tests are time-consuming, and they have limited sensitivity x-ray for the detection of covid-19 in patients. Meanwhile using an x-ray machine is a more feasible, available, and economical option. For the same reason, we consider x-ray images for our research purpose. We implemented the CNN model on the image dataset of the x-ray of the patient and successfully achieved an accuracy of 96%. Sometimes RT-PCR is a false negative in such a situation it would be helpful and avoid RT-PCR. By using modern AI techniques, the x-ray images of covid-19 patients in an automated manner can be used to diagnose at the settings where trained radiologists are not available.