ABSTRACT
Background: Automated image segmentation is an essential step in quantitative image analysis. This study assesses the performance of a deep learning-based model for lung segmentation from computed tomography (CT) images of normal and COVID-19 patients. Methods: A descriptive-analytical study was conducted from December 2020 to April 2021 on the CT images of patients from various educational hospitals affiliated with Mashhad University of Medical Sciences (Mashhad, Iran). Of the selected images and corresponding lung masks of 1,200 confirmed COVID-19 patients, 1,080 were used to train a residual neural network. The performance of the residual network (ResNet) model was evaluated on two distinct external test datasets, namely the remaining 120 COVID-19 and 120 normal patients. Different evaluation metrics such as Dice similarity coefficient (DSC), mean absolute error (MAE), relative mean Hounsfield unit (HU) difference, and relative volume difference were calculated to assess the accuracy of the predicted lung masks. The Mann-Whitney U test was used to assess the difference between the corresponding values in the normal and COVID-19 patients. P<0.05 was considered statistically significant. Results: The ResNet model achieved a DSC of 0.980 and 0.971 and a relative mean HU difference of -2.679% and -4.403% for the normal and COVID-19 patients, respectively. Comparable performance in lung segmentation of normal and COVID-19 patients indicated the model's accuracy for identifying lung tissue in the presence of COVID-19-associated infections. Although a slightly better performance was observed in normal patients. Conclusion: The ResNet model provides an accurate and reliable automated lung segmentation of COVID-19 infected lung tissue.A preprint version of this article was published on arXiv before formal peer review (https://arxiv.org/abs/2104.02042).
Subject(s)
COVID-19 , COVID-19/diagnostic imaging , Humans , Lung/diagnostic imaging , Neural Networks, Computer , Thorax , Tomography, X-Ray Computed/methodsABSTRACT
BACKGROUND: Mass screening for the coronavirus disease 2019 (COVID-19) began in Iran on March 23, 2020, with the purpose of improving early detection of patients for their own health and to prevent onward transmission to others. In this study, we evaluated the impact of the change towards mass screening on new cases reported, cases recovered, and deaths due to COVID-19. METHODS: This study analyzed the daily reports on the number of new cases confirmed by polymerase chain reaction (PCR) testing, cases recovered, and deaths due to COVID-19 provided to the Ministry of Health and Medical Education of Iran. Changes in trends on these outcomes were evaluated using interrupted time series analysis. RESULTS: From February 19 to May 6, 2020, a total of 519544 COVID-19 tests were done and 101650 diagnoses were made (case/ test ratio 19.6%). For the same period, 6418 deaths due to COVID-19 were reported (case fatality ratio 6.3%). The number of cases detected increased significantly over the period of scale-up of mass screening (P=0.003), as did the number of recovered cases (P=0.001). The number of deaths due to COVID-19 did not change before versus after mass screening. CONCLUSION: Following the scale-up of mass screening for COVID-19 in Iran, the rate of new cases detected and reported recovered accelerated significantly. Mass screening is likely to have detected many mild and asymptomatic cases that were infectious. Our data support the role that mass screening, coupled with isolation and contract tracing, can have in slowing the COVID-19 epidemic.