Development and validation of a CT-based radiomics nomogram for predicting overall survival in primary tracheal malignancy

开发和验证基于CT的放射组学列线图,用于预测原发性气管恶性肿瘤患者的总生存期

阅读:1

Abstract

OBJECTIVES: Primary tracheal malignancies (PTMs) are rare and histologically diverse, leading to complex and contentious prognostic factors. Consequently, accurately predicting the survival outcomes of these patients is challenging. This study aimed to develop a radiomics-based prognostic model for individual prediction of survival risk in PTM patients. METHODS: A total of 115 patients with PTM were reviewed retrospectively and divided into the training cohort (n = 85) and validation cohort (n = 30). Radiomics features associated with overall survival (OS) were selected using the least absolute shrinkage and selection operator (LASSO) method and combined to form the radiomics score (Radscore). Multivariable analyses were used to identify clinical and CT features as independent risk factors for OS. Radscore and identified risk factors were combined to construct a radiomics nomogram. The predictive efficacy and clinical net benefit of the prognostic models were assessed using the C-index and decision curve analysis (DCA). RESULTS: Seven radiomics features were selected by LASSO to form a Radscore. Longitudinal length was identified as an independent prognostic factor for OS. Compared with longitudinal length (C-index: 0.59; 95% confidence interval [CI]: 0.47-0.70), both the Radscore (C-index: 0.75; 95% CI, 0.63-0.87) and nomogram (C-index: 0.79; 95% CI, 0.69-0.88) demonstrated better predictive performance and were confirmed in the validation cohort. In addition, DCA indicated that both the Radscore and nomogram provided favorable clinical net benefits. CONCLUSIONS: The CT-based nomogram, which combined Radscore and longitudinal length, could individually predict the survival outcomes of PTM patients, aiding clinical decision-making.

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。