Investigating the Clinical Utility of Enhanced CT Radiomics in Distinguishing between Benign and Malignant Tumors of the Parotid Gland
Objective Create a radiomics model and a corresponding nomogram employing features derived from contrast-enhanced CT scans to classify parotid tumors as either benign or malignant.Methods A retrospective analysis was conducted on 103 patients from Shanxi Baiqiu'en Hospital who underwent two-phase contrast-enhanced CT scans of the maxillofacial region.Pathological confirmation revealed that all patients had parotid tumors(71 benign cases,32 malignant cases).Tumor regions of interest(ROIs)were carefully outlined manually on the biphasic contrast-enhanced CT scans,from which radiomic features were subsequently extracted.After feature selection using LASSO dimensionality reduction,a radiomics score(rad-score)was calculated,and a radiomics-based predictive model was developed and represented using a columnar chart.The model's performance was evaluated using ROC curves,and the model fit was evaluated using the Hosmer-Lemeshow test,and calibration curves were generated to determine the model's accuracy.Additionally,decision curve analysis(DCA)was utilized to assess the model's clinical efficacy.The Results showed that the logistic regression(LR)model,based on the radiomics score,achieved an area under the curve(AUC)of 0.848(95％CI 0.724-0.931)in the training set and an AUC of 0.826(95％CI 0.711-0.925)in the testing set.The calibration curve demonstrated good consistency,and the decision curve analysis showed high net benefits as well.Conclusion Enhanced CT radiomics effectively provides preoperative diagnostic insight for distinguishing between benign and malignant parotid tumors,thereby supporting precise clinical diagnosis and treatment planning.
万方数据
维普
