• Version
• Download 13
• File Size 649.25 KB
• File Count 1
• Create Date 7 June 2023
• Last Updated 7 June 2023

COVID-19 PREDICTION USING DENSENET-121

Harish Raj M, Harish Raja R, Arjun Ramesh R, Dr. K. Meenakshi

IStudent (B.Tech), SRM Institute of Science and Technology  Asst. Professor, SRM Institute of Science and Technology

Abstract
Coronavirus 2019, also known as COVID-19, has recently had a negative influence on public health and human lives. Since the Second World War, this disastrous effect has changed human experience by bringing about a health crisis that is unpredictable and exponentially more harmful (Kursumovic et al. in Anaesthesia 75: 989-992, 2020). The global catastrophe is a serious pandemic because to COVID19's highly contagious qualities within human groups. Early and reliable detection of the virus can be a promising method for tracking and preventing the infection from spreading (for example, by isolating the patients) due to the lack of a COVID-19 vaccine to control rather than cure the illness. This issue points to the need to improve the COVID-19 auxiliary detection method. Due of its anticipated availability, computed tomography (CT) imaging is a commonly used technology for pneumonia. The examination of photos helped by artificial intelligence may be a promising approach for finding COVID-19. In this study, convolutional neural networks (CNN) are used to provide a potential technique for predicting COVID-19 patients from the CT scan. The innovative method for forecasting COVID-19 is based on the most current modification to CNN architecture (DenseNet-121). The results exceeded 92% accuracy, and a 95% recall rate indicated that the prediction of COVID-19 performed as expected.

Keywords

DENSENET-121