Feasibility of a CT-based lymph node radiomics nomogram in detecting lymph node metastasis in PDAC patients

• DOI: 10.3389/fonc.2022.992906
• Publication Date: 2022-10-05T05:37:37Z
• Authors: Qian Li; Zuhua Song; Dan Zhang; Xiaojiao Li; Qian Liu; Jiayi Yu; Zongwen Li; Jiayan Zhang; Xiaofang Ren; Youjia Wen; Zhuoyue Tang
• Keywords: Nomogram

To investigate the potential value of a contrast enhanced computed tomography (CECT)-based radiological-radiomics nomogram combining lymph node (LN) radiomics signature and LNs' radiological features for preoperative detection LN metastasis in patients with pancreatic ductal adenocarcinoma (PDAC).In this retrospective study, 196 LNs 61 PDAC were enrolled divided into training (137 LNs) validation (59 cohorts. Radiomic extracted from portal venous phase images LNs. The least absolute shrinkage selection operator (LASSO) regression algorithm 10-fold cross-validation was used to select optimal determine score (Rad-score). developed by using significant predictors multivariate logistic (LR) analysis cohort validated independently. Its diagnostic performance assessed receiver operating characteristic curve (ROC), decision (DCA) calibration analyses.The model, including size, margin enhancement pattern (three predictors), exhibited areas under curves (AUCs) 0.831 0.756 cohorts, respectively. Nine radiomic construct which showed AUCs 0.879 0.804 nomogram, incorporated Rad-score three features, performed better than models individually, 0.937 0.851 Calibration DCA revealed that satisfactory consistency highest net benefit diagnosis metastasis.The CT-based may serve as valid convenient computer-aided tool personalized risk assessment help clinicians make appropriate clinical decisions PADC patients.