Objective To evaluate the value of conventional ultrasound,shear wave elastography and thyroglobulin antibody in the differential diagnosis of benign and malignant thyroid nodules in pilots by Logistic regression model.Methods The cohort study was used.The clinical data of pilots with thyroid nodules who were admitted to the 960th Hospital of the Joint Logistics Support Force from January of 2010 to May of 2022 were collected,including thyroglobulin antibody results and ultrasound images.The pathological results were used as the gold standard to establish a binary Logistic regression model and evaluate its ability to predict benign and malignant nodules.Results Seventy-two pilots were included,all were male,with 22-55 years old and 75-5 600 h flying hours.A total of 85 nodules were detected.Among them,48 nodules were benign nodules,including 34 nodular goiters,10 adenomas,and 4 thyroid cystic nodules.There were 37 malignant nodules,all of which were papillary carcinomas.A total of 4 statistically significant characteristic variables were screened out,including thyroid nodule morphology,microcalcification,serum thyroglobulin antibody,and maximum elastic modulus.The sensitivity,specificity and accuracy of Logistic regression model in predicting benign and malignant thyroid nodules was 83.9%,79.6%and 81.5%,respectively.The area under the receiver operating characteristic curve was 0.908(95%CI:0.850-0.966).Conclusions The Logistic regression model established by shear wave elastography combined with thyroglobulin antibody has high diagnostic value.It is helpful to the differential diagnosis of benign and malignant thyroid nodules in pilots.