To investigate the application value of TI-RADS classification for different pathological types of thyroid carcinoma.

The data of 120 patients with thyroid cancer from July 2015 to July 2018 were retrospectively analyzed. Thyroid cancer were confirmed in all patientsby ultrasonography and pathology. A total of 200 nodules were detected, including 108 malignant nodules. Statistical software SPSS22.0 was used to analyze. The results of pathological diagnosis, TI-RADS classification and different pathological types of thyroid cancer were compared by χ²(%); P<0.05 had significant difference.

Pathological results showed that 108 cases of papillary thyroid cancer, 4 cases of follicular thyroid cancer, 6 cases of medullary thyroid cancer and 2 cases of eosinophilic thyroid tumors. The diagnostic efficacy of TI-RADS classification showed that the positive likelihood ratio was 3.35 < 10, while the negative likelihood ratio was 0.13 > 0.1; the positive predictive value was 91.82%, while the negative predictive value was 75.61%. These results indicated that TI-RADS classification was effective for patients. The diagnostic efficiency of pathological type is high.

The TIRADS classification is more accurate in the diagnosis of papillary carcinoma, nodular goiter combined with papillary carcinoma, and inaccurate in the classification of medullary carcinoma, follicular carcinoma, and chronic thyroiditis. Using TI-RADS to classify thyroid pathological types can effectively improve the diagnostic accuracy of thyroid nodules, and clinically related fields can be widely promoted and applied.