Diagnostic Performance Evaluation of a Computer-Assisted Imaging Analysis System for Ultrasound Risk Stratification of Thyroid Nodules

Leesburg, VA, July 8, 2019—Although researchers at the Autonomous University of Barcelona in Badalona, Spain, led by Dr. Jordi Reverter, determined that the first commercially available system for computer-aided diagnosis (CADx) imaging analysis is useful for risk stratification of thyroid nodules, it does not perform better than a sonography expert, according to an Original Research article in the American Journal of Roentgenology.

Having received U.S. Food and Drug Administration (FDA) clearance, European Union certification, and medical device marketing approval from the Taiwan FDA, the AmCAD BioMed Corporation’s AmCAD-UT software is the first CADx system assessed and approved for human diagnosis of thyroid lesions on the basis of 2D ultrasound images. 

To evaluate CADx efficacy, the pathology reports of 300 patients who had consecutively undergone thyroidectomy at Germans Trias i Pujol Hospital in Badalona between 2015 and 2018 were retrieved. 

Of these 300 thyroid nodules, 135 were malignant, and their mean maximum nodule diameter was 3.2 ± 1.0 cm. 

For benign nodules, the mean maximum nodule diameter was 2.8 ± 0.4 cm. 

All 300 ultrasound (US) images were retrospectively reviewed by a senior staff endocrinologist with 10 years of experience in diagnostic and therapeutic thyroid US procedures who assigned each one a suspicion of malignancy score according to the American Thyroid Association (ATA). 

The AmCAD-UT software was then applied to categorize the images based on the ATA’s system, the European Thyroid Imaging Reporting and Data System, and the system jointly proposed by the American Association of Clinical Endocrinologists, American College of Endocrinology, and Associazione Medici Endocrinologi. 

The diagnostic performance of the thyroid expert when ATA classification was used was as follows: sensitivity, 87.0%; specificity, 91.2%; positive predictive value, 90.5%; negative predictive value, 90.9%. 

Compared with the expert, the AmCAD-UT, when used with the three classification systems, had similar sensitivity but lower specificity and positive predictive values. 

As for negative predictive value, the endocrinologist’s results did not differ from the AmCAD-UT when it applied the ATA classification (90.9% versus 86.3%; p = 0.07). 

The ROC AUC value was 0.88 for the clinician and 0.72 for the CADx when the ATA’s system was utilized.

Founded in 1900, the American Roentgen Ray Society (ARRS) is the first and oldest radiology society in the North America, dedicated to the advancement of medicine through the profession of radiology and its allied sciences. An international forum for progress since the discovery of the x-ray, ARRS maintains its mission of improving health through a community committed to advancing knowledge and skills with an annual scientific meeting, monthly publication of the peer-reviewed American Journal of Roentgenology (AJR), quarterly issues of InPractice magazine, topical symposia and AJR Live Webinars, print and online educational materials, as well as awarding scholarships via The Roentgen Fund®.