Feed

Lymph node metastasis is the most prevalent form of spread in thyroid cancer, with surgical intervention and prophylactic cervical lymph node dissection being the standard treatments. However, there remains a lack of effective methods for predicting lymph node metastasis in clinical practice. The study enrolled patients who were admitted to the Department of Thyroid and Neck Oncology at Tianjin Medical University Cancer Hospital between October 2021 and March 2022, as well as from February to April 2024. Variables collected included basic patient information, laboratory tests, and pathological data. Additionally, we focused on investigating the impact of emotional problems on lymph node metastasis. Subsequently, variable selection was performed using Least absolute shrinkage and selection operator (LASSO) regression and multivariate logistic regression analyses. A nomogram was then constructed for predicting lymph node metastasis, with model performance assessed using calibration curves, decision curves and other methods. In this study, a total of 484 cases were ultimately included, and the variables were screened using LASSO regression. Subsequently, nine variables were utilized to construct a nomogram. For model training purposes, 85% of the data was randomly selected as the training set and the prediction efficiency was verified using ROC curve analysis. The AUC index for both the training set (0.8045) and verification set (0.8146) were obtained along with calibration curve and decision curve plots. This study developed a nomogram to predict lymph node metastasis in papillary thyroid cancer. This model demonstrates exceptional discrimination and calibration, providing invaluable assistance for clinical decision-making processes.