Automatic Segmentation Using Deep Learning for Online Dose Optimization During Adaptive Radiotherapy of Cervical Cancer

Abstract Purpose This study investigated deep learning models for automatic segmentation to support the development of daily online dose optimization strategies, eliminating the need for ITV expansions and thereby reducing toxicity events of IMRT for cervical cancer. Methods and Materials The cervix-uterus, vagina, parametrium, bladder, rectum, sigmoid, femoral heads, kidneys, spinal cord, and bowel bag were delineated on 408 CT scans from patients treated at XXXXXXX (n=214), XXXXXXX (n=30), and enrolled in an MICCAI challenge (n=3). The data were divided into 255 training, 61 validation, 62 internal test and 30 external test CT scans. Two models were investigated: the 2D DeepLabV3+ (Google) and 3D Unet in RayStation (RaySearch Laboratories). Three IMRT plans were generated on each CT of the internal and external test sets using either the manual, 2D model, or 3D model segmentations. The dose constraints followed the EMBRACE II protocol with reduced margins of 5- and 3-mm for the target and nodal planning target volume. Geometric discrepancies between the manual and predicted contours were assessed using the Dice similarity coefficient, distance-to-agreement, and Hausdorff distance. Dosimetric discrepancies between the manual and model doses were assessed using clinical indices on the manual contours and the gamma index. Inter-observer variability was assessed for the cervix-uterus, parametrium, and vagina for the definition of the primary clinical target volume (CTVT) on the external test set. Results Average DSCs across all organs were 0.67-0.96, 0.71-0.97, and 0.42-0.92 for the 2D model and 0.66-0.96, 0.70-0.97, and 0.37-0.93 for the 3D model, on the validation, internal, and external test sets. Average DSCs of the CTVT were 0.88 and 0.81 for the 2D model and 0.87 and 0.82 for the 3D model on the internal and external test sets. Inter-observer variability of the CTVT corresponded to a mean (range) DSC of 0.85 (0.77-0.90) on the external test set. On the internal test set, the doses from the 2D and 3D model contours provided a CTVT V42.75Gy>98% for 98% and 91% of the CT scans, respectively. On the external test set, these percentages were increased to 100% and 93% for the 2D and 3D models, respectively. Conclusion The investigated models provided auto-segmentation of the cervix anatomy with similar performances on two institutional datasets and reasonable dosimetric accuracies using small PTV-margins, paving the way to automatic online dose optimization for advanced adaptive radiotherapy strategies.