AUTHOR=Mao Yi , Jiang LiPing , Wang Jing-Ling , Chen Fang-Qun , Zhang Wie-Ping , Liu Zhi-Xing , Li Chen 

TITLE=Radiomic nomogram for discriminating parotid pleomorphic adenoma from parotid adenolymphoma based on grayscale ultrasonography

JOURNAL=Frontiers in Oncology

VOLUME=Volume 13 - 2023

YEAR=2024

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2023.1268789

DOI=10.3389/fonc.2023.1268789

ISSN=2234-943X

ABSTRACT=Objectives: To differentiate parotid pleomorphic adenoma (PA) from adenolymphoma (AL) using radiomics of gray scale ultrasonography in combination with clinical features.Methods：This retrospective study aimed to analyze the clinical and radiographic characteristics of 162 cases from December 2019 to March 2023. The study population consisted of a training cohort of 113 patients and a validation cohort of 49 patients. Gray scale ultrasonography was processed using ITP-Snap software and Python to delineate regions of interest (ROIs) and extract radiomic features. Univariate analysis, Spearman correlation, greedy recursive elimination strategy, and LASSO correlation were employed to select relevant radiographic features. Subsequently, eight machine learning methods (LR, SVM, KNN, RandomForest, ExtraTrees, XGBoost, LightGBM, and MLP) were employed to build a quantitative radiomics model using the selected features. A radiomic nomogram was developed through the utilization of multivariate logistic regression analysis, integrating both clinical and radiomics data. The accuracy of the nomogram was assessed using receiver operating characteristic (ROC) curve analysis, calibration, decision curve analysis (DCA), and the Hosmer-Lemeshow test.Results：To differentiate PA and AL, the radiomic model using SVM showed optimal discriminatory ability (accuracy=0.929 and 0.857, sensitivity=0.946 and 0.800, specificity=0.921 and 0.897, positive predictive value=0.854 and 0.842, negative predictive value=0.972 and 0.867 in the training and validation cohorts, respectively), a nomogram incorporating rad-Signature and clinic features achieved an area under the ROC curve (AUC) of 0.983 (95% confidence interval [CI]: 0.965-1) and 0.910 (95% CI: 0.830-0.990) in the training and validation cohorts, respectively. Decision curve analysis showed that the nomogram and radiomic model outperformed the clinicalfactors model in terms of clinical usefulness.Conclusions：A nomogram based on gray scale ultrasonic radiomics and clinical features served as a non-invasive tool capable of differentiating PA and AL