Objective: To compare the diagnostic efficacy of the 2011 thyroid imaging reporting and data system by Kwak (Kwak TI-RADS), 2015 American Thyroid Association (ATA) management guidelines and 2017 American College of Radiology thyroid imaging reporting and data system (ACR TI-RADS) for benign and malignant thyroid nodules. Methods: The conventional ultrasound features of 265 thyroid nodules were retrospectively analyzed and the above three methods were applied to diagnose and grade the nodules. Based on the gold standard of pathological results, the receiver operating characteristic curve (ROC) was constructed. Results: The area under the ROC curve of Kwak TI-RADS, ATA guidelines grading and ACR TI-RADS classification was respectively 0.914, 0.829, and 0.910, and they all had high diagnostic efficacy for thyroid nodules, among which the diagnostic efficacy of Kwak TI-RADS and ACR TI-RADS was higher than that of ATA guidelines (P<0.05), with no significant difference between Kwak TI-RADS and ACR TI-RADS (P>0.05). The optimal diagnostic value of these three methods was respectively Kwak TI-RADS 4c, highly suspected malignancy of ATA guidelines and ACR TI-RADS 5. The accuracy of Kwak TI-RADS in the diagnosis of malignant nodules was higher than that of the other two (P<0.05). The specificity and positive predictive value of ACR TI-RADS were higher, and its sensitivity and negative predictive value were lower than those of the other two. Therefore, some thyroid nodules could not be clearly classified by ATA guidelines. Conclusion: The three methods have good efficacy in the diagnosis of thyroid nodule, and the diagnostic efficacy of Kwak TI-RADS and ACR TI-RADS was higher than that of ATA guidelines.