Predicting the nature of thyroid nodules by nomogram modeling: A study of health checkup data in a Chinese region

Background: nomogram modeling of the nature of thyroid nodules (tns) is useful in helping physical examiners to make early interventions for malignant nodules. to predict the nature of tns (benign and malignant) in a chinese population undergoing physical examination by using nomogram model. materials and methods: basic information and ultrasound (us) images were collected from 4,144 examiners who were found to have tns during their physical examinations between 2023 and 2024. predictors of malignant thyroid nodules were assessed by univariate and multivariate logistic regression. the examiners' information was randomly categorized into the training set (n = 700) and the test set (n = 300) in a 7:3 ratios. the nomogram model was constructed based on the training set, and the rocr and rms program packages were used to plot the receiver operating characteristic (roc) curve and calculate the area under curve (auc) to evaluate the classification performance of the model. results: the maximum diameter of tns (p = 0.002), waist-to-hip ratio (p = 0.002), diastolic blood pressure (p < 0.001), tsh (p < 0.001), ft4 (p < 0.001), t4 (p = 0.013), thyroglobulin (p ≤ 0.001), cea (p = 0.007), women (p = 0.012), hypertension (p = 0.047), and multiple nodules (p < 0.001) were predictors of malignant thyroid nodules. the nomogram model constructed on the basis of waist-to-hip ratio, diastolic blood pressure, maximum diameter of tns, and cea values was able to better predict malignant thyroid nodules. conclusions: our nomogram model for the nature of tns constructed on the basis of physical examination information has high accuracy, and can provide some decision support for patients with tns.