Risk Assessment Using Early Quantitative Chest CT Parameters for the Severity of COVID-19
| Author | Xun Ding |
| Author | Jia Xu |
| Author | Haibo Xu |
| Author | Jun Zhou |
| Author | Qingyun Long |
| Issued Date | 2021-07-31 |
| Abstract | Background: Today, the outbreak of coronavirus disease 2019 (COVID-19) is known as a public health emergency by the World Health Organization (WHO). Therefore, risk assessment is necessary for making a correct decision in disease management. Objectives: This study aimed to assess the risk of progression to the critical stage in COVID-19 patients, based on the early quantitative chest computed tomography (CT) parameters. Patients and Methods: In this case-control study, 39 laboratory-confirmed critical or expired COVID-19 cases (critical group), as well as 117 laboratory-confirmed COVID-19 patients including mild, moderate, and severe cases (non-critical group), were enrolled. Seven quantitative CT parameters, representing the lung volume percentages at different density intervals, were automatically calculated, using the artificial intelligence (AI) algorithms. Multivariable-adjusted logistic regression models, based on the quantitative CT parameters, were established to predict the adverse outcomes (critical vs. non-critical). The predictive performance was estimated using the receiver operating characteristic (ROC) curve analysis and by measuring the area under the ROC curve (AUC). The quantitative CT parameters in different stages were compared between the two groups. Results: No significant differences were found between the two groups regarding the lung volume percentages at different density intervals within 0 - 4 days (P = 0.596-0.938); however, this difference began to become significant within 5 - 9 days and persisted even after one month. Overall, the quantitative CT parameters could well predict the severity of COVID-19. The lung volume percentage of -7 Hounsfield units (-7 HUs) had the largest crude odds ratio (OR: 1.999; 95% CI, 1.453 ~ 2.750; P < 0.001) and adjusted OR (adjusted OR: 1.768; 95% CI, 1.114 ~ 2.808; P = 0.016). The lung volume percentage of -6 HU showed the best predictive performance with the largest AUC of 0.808; the cutoff value of 5.93% showed 71.79% sensitivity and 84.62% specificity. Conclusion: Early quantitative chest CT parameters can be measured to assess the risk of progression to the critical stage of COVID-19; this is of critical importance in the clinical management of this disease. |
| DOI | https://doi.org/10.5812/iranjradiol.109439 |
| Keyword | Quantitative CT |
| Keyword | COVID-19 |
| Keyword | Critical |
| Keyword | Artificial Intelligence |
| Publisher | Brieflands |
| Title | Risk Assessment Using Early Quantitative Chest CT Parameters for the Severity of COVID-19 |
| Type | Research Article |