The clinical value of ultrasound radiomics combined with clinical data nomogram in predicting cervical lymph node metastasis of papillary thyroid carcinoma.

Objectives: To evaluate the predictive value of a nomogram incorporating ultrasound and clinical features for cervical lymph node metastasis (CLNM) in patients with thyroid cancer. Methods: A retrospective analysis was conducted on patients pathologically diagnosed with papillary thyroid carcinoma (PTC) between April 2024 and January 2025. Radiomics features were extracted from two-dimensional thyroid ultrasound images, and feature selection was performed using the least absolute shrinkage and selection operator (LASSO). Five classifiers were compared, and logistic regression was selected to construct the radiomics model. A clinical-radiomics model integrating the radiomics score and clinical predictors, together with a standalone clinical model, was developed using multivariable logistic regression. Model discrimination, calibration, and clinical utility were evaluated in a held-out internal validation set. Results: Sex, Radscore, age, and thyroid stimulating hormone (TSH) levels were identified as independent predictors of CLNM in patients with PTC. Based on these variables, radiomics, clinical, and clinical radiomics models were constructed; the models achieved area under the curve (AUC) values of 0.735, 0.792, and 0.867, respectively, in the validation set. DeLong's test and decision curve analysis (DCA) demonstrated that the clinical radiomics model outperformed the radiomics model in the validation set, offering a higher net clinical benefit. Conclusion: A nomogram integrating ultrasound imaging and clinical data effectively predicts the risk of CLNM in patients with PTC and may assist clinicians in developing individualized treatment strategies.