Background: The purpose of this study was to compare the diagnostic efficiency of a thyroid ultrasound computer-aided diagnosis (CAD) system with that of 1 radiologist.
Methods: This study retrospectively reviewed 342 surgically resected thyroid nodules from July 2013 to December 2013 at our center. The nodules were assessed on typical ultrasound images using the CAD system and reviewed by 1 experienced radiologist. The radiologist stratified the risk of malignancy using the Thyroid Imaging Reporting and Data Systems (TIRADS) and the American Thyroid Association (ATA) guidelines.
Results: The radiologist, using TI-RADS and ATA guidelines, performed better than the CAD system (P < .01). The sensitivity of the CAD system was similar to that of an experienced radiologist (P > .05; P < .01; and P > .05). However, we found that the CAD system had lower specificity (P < .01).
Conclusion: The sensitivity of a thyroid ultrasound CAD system in differentiating nodules was similar to that of an experienced radiologist. However, the CAD system had lower specificity.
Keywords: 2015 American Thyroid Association management guideline; Thyroid Imaging Reporting and Data System; artificial intelligence; computer-aided diagnosis system; thyroid nodules; ultrasound.
© 2017 Wiley Periodicals, Inc.