【】Objective?? Explore the ultrasound characteristics of thyroid tumors and the value of constructing a nomogram prediction model.?? Methods?? We selected 70 patients with thyroid malignant nodules confirmed by pathological examination in our hospital from January 2019 to June 2023 as the malignant group, and 90 patients with thyroid benign nodules confirmed at the same time as the benign group. We compared the ultrasound characteristics of the nodules, thyroid ultrasound imaging reports and data system (TI-RADS) classification, general information, and plotted column charts for statistically significant indicators between the two groups, Establish a predictive model for ultrasound diagnosis of benign and malignant thyroid nodules.?? Results?? The proportion of low echogenicity, microcalcification detection rate, aspect ratio≥1, halo thickness≥2 mm, Adler grade≥2, and posterior echo attenuation of thyroid nodules in malignant group patients were significantly higher than those in benign group patients, and the differences were statistically significant (P<0.05); There was no statistically significant difference in the shape, edge features, and number of nodules between the benign and malignant groups of patients (P>0.05); The proportion of patients with TI-RADS classification of 4, 5, and≥3 categories in thyroid nodules in the malignant group was significantly higher than that in the benign group, and the differences were statistically significant (P<0.05); The logistic regression model showed that hypoechogenicity, aspect ratio≥1, halo thickness≥2 mm, Adler grade≥2, and TI-RADS grade ≥3 were independent risk factors for thyroid malignant nodules (P<0.05); The TI-RADS classification has the greatest contribution in column chart analysis, followed by the aspect ratio of nodules, halo thickness, Adler grading, and nodule echo characteristics. The column chart model has good diagnostic value (C index=0.774, 95% CI: 0.628~0.811); Using pathological results as the gold standard, the AUC value under the ROC curve for TI-RADS classification and differential diagnosis of benign and malignant thyroid nodules is 0.810, with a sensitivity of 78.83% and a specificity of 80.64%; The AUC value of the ROC area under the curve for distinguishing benign and malignant thyroid nodules using the ultrasound characteristic column chart model is 0.891, with a sensitivity of 91.73% and a specificity of 84.33%. The diagnostic efficacy of the column chart is higher.Conclusion Establishing a line graph model based on thyroid ultrasound features, classified according to TI-RADS, followed by the ratio of longitudinal and transverse diameters of nodules, halo thickness, Adler grading, and nodule echo characteristics, has high value in the comprehensive analysis and diagnosis of benign and malignant thyroid nodules.