Related Experiment Video
Updated: May 14, 2026
Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
Published on: April 18, 2025
A Practical Nomogram for Preoperative Prediction of Lateral Lymph Node Metastasis in Children and Adolescent Patients
Xiaoqing Wang1, Ying Li2, Jingtong Li1
1Department of Diagnostic and Therapeutic Ultrasonography, Tianjin Medical University Cancer Institute & Hospital, National Clinical Research Center of Cancer, Tianjin's Clinical Research Center for Cancer, Tianjin Key Laboratory of Basic and Translational Medicine on Head & Neck Cancer, Tianjin, China.
Background:
Lateral lymph node metastasis (LLNM) is an independent prognostic factor associated with poor outcomes in pediatric patients with papillary thyroid carcinoma (PTC). This study aims to develop and validate an effective preoperative nomogram that integrates clinical and ultrasound features to predict LLNM in children and adolescents with PTC.
Methods:
This study included 300 pediatric patients diagnosed with PTC who underwent either total or partial thyroidectomy at three medical centers between January 2013 and March 2025. The subjects were divided into three datasets for training, internal testing and external testing. We constructed a comprehensive nomogram that incorporated both clinical and ultrasound features and employed a multivariate logistic regression model to predict LLNM in children and adolescents with PTC. The model was evaluated in terms of discrimination, calibration and clinical utility.
Results:
The training, internal testing and external testing datasets comprised 177, 75 and 48 pediatric patients with PTC, respectively. Multivariate logistic regression analysis revealed that vascularization, the relationship to the thyroid capsule, age and multifocality were independent risk factors for predicting LLNM in the training dataset. The predictive nomogram model developed using these variables demonstrated strong performance in estimating the risk of LLNM, with an area under the curve of 0.779 (95% confidence interval [CI], 0.711-0.847) for the nomogram in the training cohort, 0.789 (95% CI, 0.678-0.899) in the internal testing cohort, and 0.933 (95% CI, 0.865-1.000) in the external testing cohort.
Conclusion:
Our research findings indicate that the newly developed nomogram, which is based on clinical and ultrasound characteristics, effectively predicts LLNM in children and adolescents with PTC preoperatively. This tool may assist clinicians in selecting appropriate surgical strategies.
