Evaluating the Effectiveness of Using Artificial Intelligence to Detect COVID-19

Authors

Adam Fritz, University of LynchburgFollow

Specialty

Infectious Disease

Advisor

Dr. Nancy Reid

Abstract

Objectives: To evaluate the effectiveness of the use of artificial intelligence and machine learning to detect the Coronavirus disease 2019 (COVID-19) in chest radiography.

Methods: In this retrospective review data was collected from previously published literature to evaluate the accuracy of detecting COVID-19 on chest radiography with the use of artificial intelligence. Studies with larger datasets (>1000 images) were targeted for review.

Results: The application of convolutional neural networks and deep learning models with varying architecture produced results achieving 93.3%-99.1% accuracy for the binary category.

Conclusions: Review of the literature revealed strong evidence that artificial intelligence can expedite the identification process of COVID-19 through chest abnormalities and relieve strain on other medical resources. However, this technology needs to have an established standardized model and greater dissemination in order to increase its potential impact on medical systems and resources.

Recommended Citation

Fritz A. Evaluating the Effectiveness of Using Artificial Intelligence to Detect COVID-19. University of Lynchburg DMSc Doctoral Project Assignment Repository. 2021; 3(4).