Development and validation of a CT-texture analysis nomogram for preoperatively differentiating thymic epithelial tumor histologic subtypes

Abstract

Background Thymic epithelial tumors (TETs) are the most common primary tumors in the anterior mediastinum, which have considerable histologic heterogeneity. This study aimed to develop and validate a nomogram based on computed tomography (CT) and texture analysis (TA) for preoperatively predicting the pathological classifications for TET patients. Methods Totally TET 172 patients confirmed by postoperative pathology between January 2011 to April 2019 were retrospectively analyzed and randomly divided into training (n = 120) and validation (n = 52) cohorts. Preoperative clinical factors, CT signs and texture features of each patient were analyzed, and prediction models were developed using the least absolute shrinkage and selection operator (LASSO) regression. The performance of the models was evaluated and compared by the area under receiver-operator characteristic (ROC) curve (AUC) and the DeLong test. The clinical application value of the models was determined via the decision curve analysis (DCA). Then, a nomogram was developed based on the model with the best predictive efficiency and clinical utility and validated using the calibration plots. Results Totally 87 patients with low-risk TET (LTET) (types A, AB, B1) and 85 patients with high-risk TET (HTET) (types B2, B3, C) were enrolled in this study. We separately constructed 4 prediction models for differentiating LTET from HTET using clinical, CT, texture features, and their combination. These 4 prediction models achieved AUCs of 0.66, 0.79, 0.82, 0.88 in the training cohort and 0.64, 0.82, 0.86, 0.94 in the validation cohort, respectively. The DeLong test and DCA showed that the Combined model, consisting of 2 CT signs and 2 texture parameters, held the highest predictive efficiency and clinical utility (p < 0.05). A prediction nomogram was subsequently developed using the 4 independently risk factors from the Combined model. The calibration curves indicated a good consistency between the actual observations and nomogram predictions for differentiating TET classifications. Conclusion A prediction nomogram incorporating both the CT and texture parameters was constructed and validated in our study, which can be conveniently used for the preoperative individualized prediction of the simplified histologic subtypes in TET patients.

Computed tomography radiomics for the prediction of thymic epithelial tumor histology, TNM stage and myasthenia gravis

“…In contrast, the RF classifier in our study was able to distinguish LRT from high-risk TET (HRT) with good discriminatory performance (AUC 0.88). The performance was similar to the findings of Hu et al, who previously explored the use of CT-based radiomics derived from unenhanced and contrast-enhanced scans and machine learning classifiers for the same task and reached an AUC of 0.81 (RF classifier) [ 15 ], as well as to the findings of Ren et al, who used logistic regression to build a nomogram (AUC 0.86) [ 16 ]. Our study therefore supports the thought that quantitative, radiomic assessment of the tumor composition may outperform visual assessment of heterogeneity and shape and may enrich TET diagnostics.…”

“…Interestingly, the shape feature sphericity appeared in multiple previous studies, using different means of feature selection, as an important feature for TET characterization [ 13 , 16 , 21 , 40 , 41 ]. In our study, sphericity was among the most important features for predictions of the histologic subtype as well as the TNM stage.…”

Optimizing the radiomics-machine-learning model based on non-contrast enhanced CT for the simplified risk categorization of thymic epithelial tumors: A large cohort retrospective study

“…This should be endorsed the potential use of radiomics as a valuable tool in the individualized assessment and treatment decision-making of TETs. It is also extensively applied in the field of simplified risk categorization of TETs based on CT images [18][19][20][21][22][23]. However, a solid conclusion is far from drew suffered from the following two reasons.…”

“…First, all these studies are conducted with the small sample size (<200 patients in all studies). Second, the study protocol was not unified, making the outcome incomparable, for example, some studies are carried out using contrast-enhanced CT (CECT) [19,21], some compare only the radiomic contributions (leaving the clinical features ignored) [20,22,23], some focus on only classifying LRTs from HRTs [19][20][21]23], some use only cross-validation or internal validation to investigate the diagnostic performance [20,22], some analyses are based on 2-D regions of interest (ROIs) instead of 3-D volumes of interest (VOIs) [20], etc. Thus, it is far from conclusive on the diagnostic performance of NECT radiomics on simplified TETs risk categorization.…”

Contrast-enhanced CT-based radiomics model for differentiating risk subgroups of thymic epithelial tumors

“…The subsequent quantitative analysis of these data can offer help in differential diagnosis, risk classification, predicting prognosis and efficacy evaluation of tumors based on different kinds of medical images [ 12 – 15 ]. Although several CT-based radiomics analyses have been used to identify the risk classification of thymic epithelial tumors, most studies were based on two-classification [ 16 , 17 ]. Only one study was based on triple classification, and the accuracy of the clinical-semantic radiomics model (RM) in the risk assessment of three subgroups in the validation group was only 48.3% [ 18 ].…”