Fully automated multiorgan segmentation of female pelvic magnetic resonance images with coarse‐to‐fine convolutional neural network. Issue 11 (21st October 2021)

• Record Type: Journal Article
• Authors: Zabihollahy, Fatemeh; Viswanathan, Akila N; Schmidt, Ehud J; Morcos, Marc; Lee, Junghoon
• Is Part Of: Medical physics. Volume 48:Issue 11(2021)
• Page Start: 7028
• Page End: 7042
• Publication Date: 2021-10-21
• Subjects: deep learning -- magnetic resonance imaging -- multiorgan segmentation -- radiotherapy; Medical physics -- Periodicals; Medical physics; Geneeskunde; Natuurkunde; Toepassingen; Biophysics; Periodicals; Periodicals; Electronic journals; 610.153
• DOI: 10.1002/mp.15268
• Languages: English
• ISSNs: 0094-2405

Abstract

Abstract: Purpose: Brachytherapy combined with external beam radiotherapy (EBRT) is the standard treatment for cervical cancer and has been shown to improve overall survival rates compared to EBRT only. Magnetic resonance (MR) imaging is used for radiotherapy (RT) planning and image guidance due to its excellent soft tissue image contrast. Rapid and accurate segmentation of organs at risk (OAR) is a crucial step in MR image‐guided RT. In this paper, we propose a fully automated two‐step convolutional neural network (CNN) approach to delineate multiple OARs from T2‐weighted (T2W) MR images. Methods: We employ a coarse‐to‐fine segmentation strategy. The coarse segmentation step first identifies the approximate boundary of each organ of interest and crops the MR volume around the centroid of organ‐specific region of interest (ROI). The cropped ROI volumes are then fed to organ‐specific fine segmentation networks to produce detailed segmentation of each organ. A three‐dimensional (3‐D) U‐Net is trained to perform the coarse segmentation. For the fine segmentation, a 3‐D Dense U‐Net is employed in which a modified 3‐D dense block is incorporated into the 3‐D U‐Net‐like network to acquire inter and intra‐slice features and improve information flow while reducing computational complexity. Two sets of T2W MR images (221 cases for MR1 and 62 for MR2) were taken with slightly different imaging parameters and used for our network training and test. The network was first trained on MR1 which was a larger sample set. The trained model was then transferred to the MR2 domain via a fine‐tuning approach. Active learning strategy was utilized for selecting the most valuable data from MR2 to be included in the adaptation via transfer learning. Results: The proposed method was tested on 20 MR1 and 32 MR2 test sets. Mean ± SD dice similarity coefficients are 0.93 ± 0.04, 0.87 ± 0.03, and 0.80 ± 0.10 on MR1 and 0.94 ± 0.05, 0.88 ± 0.04, and 0.80 ± 0.05 on MR2 for bladder, rectum, and sigmoid, respectively. Hausdorff distances (95th percentile) are 4.18 ± 0.52, 2.54 ± 0.41, and 5.03 ± 1.31 mm on MR1 and 2.89 ± 0.33, 2.24 ± 0.40, and 3.28 ± 1.08 mm on MR2, respectively. The performance of our method is superior to other state‐of‐the‐art segmentation methods. Conclusions: We proposed a two‐step CNN approach for fully automated segmentation of female pelvic MR bladder, rectum, and sigmoid from T2W MR volume. Our experimental results demonstrate that the developed method is accurate, fast, and reproducible, and outperforms alternative state‐of‐the‐art methods for OAR segmentation significantly ( p < 0.05).