摘要

Background: Malignancy is correlated with stiffness which is assessed by real time elastosonography (RTE). RTE is not used in routine practice. We aimed to establish the learning curve of RTE on radiology residents. Methods: Forty >= 1 cm solitary thyroid nodules referred to fine needle aspiration cytology were examined with RTE by a radiology specialist and two radiology residents separately. Patients with malignant and undetermined FNAC findings underwent surgery. Strain ratio and elasticity score results of the radiology residents were compared to the results of the radiology specialist taking the histopathology results as the reference. To establish the learning curve and compare the diagnostic accuracy of residents, Receiver Operating Characteric curves were generated and Area Under the Curve were calculated. Results: Thirty-four nodules were benign and the others were malignant. The radiology specialist had a very high correct class prediction for the differentiation of benign versus malignant thyroid nodules. Statistical analysis of the strain ratio measurements showed that one of the residents had similar results with the radiology specialist after the seventh patient and the other one after the fourth patient. On the other hand the elasticity score measurements of all examiners had low correct class prediction. Conclusions: Strain ratio measurement by RTE is an easily learned sonographic method that can assist in the evaluation of benign versus malignant nature of the lesions. However, interpretation of the elasticity scores requires more expertise. The results of this preliminary study need to be verified with a larger sample population.

  • 出版日期2013-12