The Role of Artificial Intelligence Based Models in The Diagnosis of Covid-19 On Computed Tomography-A Systematic Review

Abstract

The main objective of this study is to emphasize the role of Artificial Intelligence based models in the diagnosis, differentiation, evaluation of the severity of COVID-19 while utilizing CT as imaging modality AI based models. This review was according to guidelines of the Preferred Reporting Items for Systematic Review and Meta-analysis (PRISMA). Fifteen articles were reviewed for the evaluation of the results of AI based models in diagnosis and in assessment of the severity of COVID-19. The presented models shows higher accuracy, sensitivity and specificity. The presented models diagnosis the virus in rapid way and it also reduces the workload on radiologists. The models are categorized in differentiating the patients with normal person from COVID-19, other pneumonia and the affected part of lungs by COVID-19.Various tests are performed for the robustness of these models, and the comparison of the AI based models results with experienced radiologists that makes them more useful and accurate in diagnostic approach. COVID-19 is continuously spreading across the World. Advanced classifying and diagnostic models for COVID-19 based on AI are widely available and proved favorable in the diagnosis of COVID-19. The models based on “deep learning” “machine learning” “multiple instance learning” and “transfer learning” are available for the diagnosis of this disease and differentiating it from others pneumonia.The sensitivities and the specificities are greater than 85%, which makes them a valuable tool for diagnosis.By improving the geographical information and number of patients for the training, can make these models very fast diagnostic tools for COVID-19.