Prospective Evaluation of Automated Contouring for CT-Based Brachytherapy for Gynecologic Malignancies
Abigayle Kraus, Zohaib Iqbal, R Cardan, Richard A. Popple, Dennis N. Stanley, Sui Shen, Joel A. Pogue, Xingen Wu, Kevin Lee, Samuel Marcrom, Carlos Cárdenas
Abstract
PurposeThe use of deep learning to auto-contour organs-at-risk (OARs) in gynecologic radiation treatment is well-established. Yet, there is limited data investigating the prospective use of auto-contouring in clinical practice. In this study, we assess the accuracy and efficiency of auto-contouring OARs for CT-based brachytherapy treatment planning of gynecologic malignancies.Materials and MethodsAn in-house contouring tool automatically delineated five OARs in gynecologic radiation treatment planning: the bladder, small bowel, sigmoid, rectum, and urethra. Accuracy of each auto-contour was evaluated using a 5-point Likert scale: a score of five indicated the contour could be used without edits, while a score of one indicated the contour was unusable. During scoring, automated contours were edited and subsequently used for treatment planning. Dice similarity coefficient (DSC), mean surface distance (MSD), 95% Hausdorff Distance (95HD), Hausdorff Distance (HD), and dosimetric changes between original and edited contours were calculated. Contour approval time and total planning time of a prospective auto-contoured (AC) cohort were compared to times from a retrospective manually contoured (MC) cohort.ResultsThirty AC cases from January 2022 to July 2022 and thirty-one MC cases from July 2021 to January 2022 were included. The mean (±SD) Likert score for each OAR was the following: bladder 4.77 (±0.58), small bowel 3.96 (±0.91), sigmoid colon 3.92 (±0.81), rectum 4.6 (±0.71), and urethra 4.27 (±0.78). No auto-contours required major edits. All OARs had a mean DSC > 0.86, MSD < 0.48mm, 95HD < 3.2mm, and HD < 10.32mm between original and edited contours. There was no significant difference in DVH metrics (D2.0cc/D0.1cc) between original and edited contours (p-values > 0.05). The average time to plan approval in the AC cohort was 19% less than the MC cohort. (AC vs. MC, 117.0 + 18.0 minutes versus 144.9 ± 64.5 minutes, p=0.045).ConclusionsAutomated contouring is useful and accurate in clinical practice. Auto-contouring OARs streamlines radiation treatment workflows and decreases time required to design and approve gynecologic brachytherapy plans.