Analysis of Axial CT-Scan image of COVID-19 patients based in gender using the Otsu Thresholding method

Abstract

At the beginning of 2020, the world was shocked by the emergence of the COVID-19 virus. This virus has spread to all corners of the world, not only in Indonesia. Therefore, the government needs to make efforts to break the chain of transmission of this virus. One of these efforts is to detect COVID-19 as early as possible. Using CT images can be one of the early detection efforts of early-phase lung infections in COVID-19 patients. The stage in detecting COVID-19 is by segmenting the image. In this study, segmentation was carried out using the Otsu Thresholding method on 8 axial CT images of the lungs of COVID-19 patients, consisting of 4 images of male patients and 4 images of female patients. Then the image segmentation results of male and female patients were compared and evaluated using ROC measurements, Threshold (T) values and analyzed for GGO (grand-glass opacity). The result can be seen that judging from the value of the ROC measurement results, the measurement of image segmentation evaluation of male patients is more accurate than female patients. The number of false negatives for male patients and female patients is the same, while the number of false positives for male patients is less than female patients. Threshold value of the image segmentation results of male and female patients is the same so that the density of image segmentation is the same. GGO (grand-glass opacity) for male COVID-19 patients aged between 45-55 years is fuller than female COVID-19 patients aged 45-55 years. This shows that men are more at risk of dying from COVID-19 than female.©2020 JNSMR UIN Walisongo. All rights reserved.