Zhou H, et al. - A highly automatic and objective model named deep learning Radiomics of thyroid (DLRT) was proposed for the differential diagnosis of benign and malignant thyroid nodules from ultrasound (US) images. A retrospective study was conducted to finally include US images and fine-needle aspiration biopsies from 1734 patients with 1750 thyroid nodules. Researchers applied a basic convolutional neural network (CNN) model, a transfer learning (TL) model, and a newly designed model named deep learning Radiomics of thyroid (DLRT) for the investigation. They conducted analysis of receiver operating characteristic (ROC) curves to estimate optimal area under it (AUC) for benign and malignant nodules. Comparing with other deep learning models and human observers, DLRT exhibits the best overall performance. Results indicate that it is a promising tool for improving the differential diagnosis of benign and malignant thyroid nodules.