The aim of this study was to develop a deep learning model for automated delineation of the clinical target volume (CTV) for cervical cancer adjuvant radiotherapy after radical hysterectomy.A total of 263 planning-computed tomography (CT) scans of postoperative radiotherapy in cervical cancer patients after radical hysterectomy treated at our institution were used to train and test the automatic contouring models. Among these sets, 236 were used for training, 27 for testing. In the study, three models were investigated: CMU-Net, V-net (3D neural network) and Unet++ (2D neural network). Multi-U-Net (CMU-Net), a novel segmentation approach, recognized the upper and lower bounds of CTV and then segmented them accurately. Dice similarity coefficient (DSC), Hausdorff distances (HD), Average symmetric surface distance (ASD) were assessed and compared between models and manual contours for testing sets.The mean DSC, HD and ASD of V-Net, Unet++, CMU-Net were 86.39, 85.74, 92.16; 5.97, 5.35, 2.72; 0.89, 0.76, and 0.52, respectively.Among the three models, CMU-Net provided the best in automatic contouring of the CTV for cervical cancer adjuvant radiotherapy, and could help physicians to further reduce the time cost in the clinical practice.Copyright © 2021. Published by Elsevier Inc.