Clinical target volume delineation quality assurance for MRI-guided prostate radiotherapy using deep learning with uncertainty estimation

• Published in: Radiotherapy and oncology 2023-09, Vol.186, p.109794-109794, Article 109794
• Main Authors: Min, Hang, Dowling, Jason, Jameson, Michael G, Cloak, Kirrily, Faustino, Joselle, Sidhom, Mark, Martin, Jarad, Cardoso, Michael, Ebert, Martin A, Haworth, Annette, Chlap, Phillip, de Leon, Jeremiah, Berry, Megan, Pryor, David, Greer, Peter, Vinod, Shalini K., Holloway, Lois
• Format: Article
• Language: English
• Subjects: Clinical target volume; Deep Learning; Humans; Magnetic Resonance Imaging; Male; Prostate cancer; Prostatic Neoplasms - diagnostic imaging; Prostatic Neoplasms - radiotherapy; Quality assurance; Quality Assurance, Health Care; Radiotherapy; Radiotherapy, Image-Guided; Uncertainty; Uncertainty estimation
• ISSN: 0167-8140; 1879-0887
• DOI: 10.1016/j.radonc.2023.109794

•A clinical target volume delineation quality assurance (QA) framework for MRI-guided prostate radiotherapy was proposed.•The framework adopted a deep learning network with Monte Carlo dropout for uncertainty estimation.•Spatial correlation between the manual delineation and the area of uncertainty generated by the network was incorporated in the QA criteria.•The framework was evaluated on a multicentre MRI-only prostate radiotherapy trial dataset and achieved promising performance in QA classification. Previous studies on automatic delineation quality assurance (QA) have mostly focused on CT-based planning. As MRI-guided radiotherapy is increasingly utilized in prostate cancer treatment, there is a need for more research on MRI-specific automatic QA. This work proposes a clinical target volume (CTV) delineation QA framework based on deep learning (DL) for MRI-guided prostate radiotherapy. The proposed workflow utilized a 3D dropblock ResUnet++ (DB-ResUnet++) to generate multiple segmentation predictions via Monte Carlo dropout which were used to compute an average delineation and area of uncertainty. A logistic regression (LR) classifier was employed to classify the manual delineation as pass or discrepancy based on the spatial association between the manual delineation and the network’s outputs. This approach was evaluated on a multicentre MRI-only prostate radiotherapy dataset and compared with our previously published QA framework based on AN-AG Unet. The proposed framework achieved an area under the receiver operating curve (AUROC) of 0.92, a true positive rate (TPR) of 0.92 and a false positive rate of 0.09 with an average processing time per delineation of 1.3 min. Compared with our previous work using AN-AG Unet, this method generated fewer false positive detections at the same TPR with a much faster processing speed. To the best of our knowledge, this is the first study to propose an automatic delineation QA tool using DL with uncertainty estimation for MRI-guided prostate radiotherapy, which can potentially be used for reviewing prostate CTV delineation in multicentre clinical trials.