Lin Z, Cui Y, Liu J, et al. - Researchers sought to construct a 3D (three dimensional) U-Net-based deep learning model for automated segmentation of kidney and renal mass, as well as identification of renal mass in the corticomedullary phase of computed tomography urography (CTU). To learn and assess the deep learning model, experts obtained data on 882 kidneys collected from CTU data of 441 patients having renal mass. Based on a 3D U-Net, the proposed segmentation model for kidney and renal mass was developed. Via the Dice similarity coefficient (DSC), assessment of the segmentation accuracy was done. A high accuracy in segmentation of kidney and renal tumor was demonstrated by the proposed model, with average DSC estimated to be 0.973 and 0.844, respectively. A good performance in detecting renal tumor and cyst was also displayed by this model. Overall, promising results for the segmentation of kidney and renal tumor as well as for the detection of renal tumor and cyst were demonstrated by the proposed automated segmentation and detection model based on 3D U-Net.