COVID-19 diagnosis in CT images using CNN to extract features and multiple classifiers

Coronavirus disease (COVID-19) has already infected more than 20 million people worldwide and is responsible for more than 744,000 deaths. A major problem faced in the diagnosis of COVID-19 is the inefficiency and scarcity of medical tests. The use of computed tomography (CT) has shown promise in the evaluation of patients with suspected COVID-19 infection. The analysis of the CT examination is complex and requires the effort of a specialist, which can lead to diagnostic errors. The use of CAD systems can minimize the problems generated by the analysis of CTs by specialists. This article presents a methodology for diagnosing COVID-19 using a trainable resource extractor using CNN and multiple classifiers. First, the quality of the images was improved using histogram equalization and CLAHE. Then, a basic CNN is used to extract resources from 708 CTs, 312 with COVID-19, and 396 Non-COVID-19. After the extracted data, we used multiple classifiers for classification in COVID-19 and Non-COVID-19. The results show an accuracy of 97.88%, recall of 97.77%, the precision of 97.94%, F-score of 0.978, AUC of 0.977, and kappa index of 0.957. The results obtained show that the proposed methodology can be used as a CAD system to aid in the diagnosis of COVID-19.