Development and validation of a diagnostic model for thoracic aortic dissection based on radiomics derived from non-enhanced computed tomography imaging

基于非增强计算机断层扫描成像的放射组学,开发和验证胸主动脉夹层诊断模型

阅读:1

Abstract

BACKGROUND: Thoracic aortic dissection (TAD) is a life-threatening disease, but the current conventional diagnostic methods are limited. In this study, we developed and validated a nomogram based on clinical features and non-enhanced computed tomography (CT) radiomics for the diagnosis of TAD. METHODS: A retrospective analysis of the data of 265 patients who underwent both non-enhanced and enhanced chest CT scans at two medical centers was conducted. The patients were randomly divided into a training set (n=147) and a validation set (n=63) at a 7:3 ratio. Patients from another institution were designated as the external test set (n=55). A multivariate logistic regression analysis was conducted to identify the independent clinical predictors and CT features. The least absolute shrinkage and selection operator (LASSO) logistic regression algorithm was used to select the optimal radiomic features. RESULTS: Combining the clinical independent predictors with the radiomics score (Rad score) enhanced the predictive power of the nomogram, which showed good consistency between the training and validation sets. In the training set, the nomogram had an area under the curve (AUC) of 0.968 [95% confidence interval (CI): 0.942-0.993], and an accuracy, specificity, and sensitivity of 92.5%, 95.9%, and 89.2%, respectively; while in the validation set, it had an AUC of 0.965 (95% CI: 0.919-1.000), and an accuracy, specificity, and sensitivity of 92.1%, 91.7%, and 92.6%, respectively. CONCLUSIONS: The nomogram based on radiomic features and clinical characteristics from non-enhanced CT imaging can effectively diagnose TAD and could serve as a tool for screening high-risk patients with TAD.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。