Deep convolutional neural network for classification of thyroid nodules on ultrasound: Comparison of the diagnostic performance with that of radiologists

•On ultrasound images, deep learning-trained models demonstrated comparable diagnostic performance to radiologists in differentiating malignant from benign thyroid nodules.•VGG16 model showed the best diagnostic performance in internal (AUC, 0.86; sensitivity, 91.8%; specificity, 73.2%) and external (AUC: 0.83; sensitivity: 78.6%; specificity: 76.8%) test sets.•Deep learning models may help radiologists’ diagnosis of thyroid nodules on ultrasound.