Feasibility of clinical target volume (CTV) automatic delineation using deep learning network for cervical cancer radiotherapy: a study with external validation

Abstract

Abstract Purpose To explore the accuracy and feasibility of a proposed deep learning (DL) algorithm for clinical target volume (CTV) delineation in cervical cancer radiotherapy and evaluate whether it can perform well to external cervical cancer and endometrial cancer cases for generalization validation. Materials and methods A total of 332 patients were enrolled in this study. A state-of-the-art network called AttResCNet, which adopted Resnet50 based on a channel and spatial attention as backbone was proposed. 236 cervical cancer cases were randomly grouped into training (n = 189) and internal validation (n = 47) cohorts. External validations were performed in a separate cohort of 54 cervical cancer and 42 endometrial cancer cases. The performances of the proposed network were evaluated by dice similarity coefficient (DSC), sensitivity (SEN), positive predictive value (PPV), 95% Hausdorff distance (95HD) and oncologist clinical score when comparing them with manual delineation in validation cohorts. Results In internal validation cohorts, the DSC, SEN, PPV, 95HD for AttResCNet achieved 0.775, 0.796, 0.740, 10.156 mm. In external independent validation cohorts, AttResCNet achieved 0.765, 0.805, 0.729, 12.075 mm for cervical cancer cases and 0.766, 0.803, 0.738, 11.527 mm for endometrial cancer cases, respectively. The clinical assessment score showed that minor and no revisions (delineation time was shortened to within 30 minutes) accounted for about 85% of all cases in DL-aided automatic delineation. Conclusions We demonstrated the problem of model generalizability for DL-based automatic delineation. The proposed network can improve the performance at automatic delineation for cervical cancer and shorten manual delineation time at no expense of quality. The network showed excellent clinical viability, which can also be even generalized for endometrial cancer with excellent performance.