Synthetic MRI-aided multi-organ segmentation on male pelvic CT using cycle consistent deep attention network

• Published in: Radiotherapy and oncology Vol. 141; pp. 192 - 199
• Main Authors: Dong, Xue, Lei, Yang, Tian, Sibo, Wang, Tonghe, Patel, Pretesh, Curran, Walter J., Jani, Ashesh B., Liu, Tian, Yang, Xiaofeng
• Format: Journal Article
• Language: English
• Published: Ireland Elsevier B.V 01.12.2019
• Subjects: Deep Learning; Hematology, Oncology, and Palliative Medicine; Humans; Magnetic Resonance Imaging - methods; Male; Multi-organ segmentation; Organs at Risk; Pelvis - diagnostic imaging; Prostatic Neoplasms - diagnostic imaging; Prostatic Neoplasms - radiotherapy; Synthetic MRI; Tomography, X-Ray Computed - methods; Deep learning; Synthetic MRI; Multi-organ segmentation
• ISSN: 0167-8140; 1879-0887; 1879-0887
• DOI: 10.1016/j.radonc.2019.09.028

•We use CT-based synthetic MRI (SMRI) to aid multi-organ segmentation in male pelvic CT.•SMRI with a superior soft-tissue contrast can significantly improve CT segmentation accuracy.•The deep attention mechanism helps identify the most relevant features to differentiate organs.•The proposed SMRI-aided strategy has a great potential to facilitate radiotherapy planning. Manual contouring is labor intensive, and subject to variations in operator knowledge, experience and technique. This work aims to develop an automated computed tomography (CT) multi-organ segmentation method for prostate cancer treatment planning. The proposed method exploits the superior soft-tissue information provided by synthetic MRI (sMRI) to aid the multi-organ segmentation on pelvic CT images. A cycle generative adversarial network (CycleGAN) was used to estimate sMRIs from CT images. A deep attention U-Net (DAUnet) was trained on sMRI and corresponding multi-organ contours for auto-segmentation. The deep attention strategy was introduced to identify the most relevant features to differentiate different organs. Deep supervision was incorporated into the DAUnet to enhance the features’ discriminative ability. Segmented contours of a patient were obtained by feeding CT image into the trained CycleGAN to generate sMRI, which was then fed to the trained DAUnet to generate organ contours. We trained and evaluated our model with 140 datasets from prostate patients. The Dice similarity coefficient and mean surface distance between our segmented and bladder, prostate, and rectum manual contours were 0.95 ± 0.03, 0.52 ± 0.22 mm; 0.87 ± 0.04, 0.93 ± 0.51 mm; and 0.89 ± 0.04, 0.92 ± 1.03 mm, respectively. We proposed a sMRI-aided multi-organ automatic segmentation method on pelvic CT images. By integrating deep attention and deep supervision strategy, the proposed network provides accurate and consistent prostate, bladder and rectum segmentation, and has the potential to facilitate routine prostate-cancer radiotherapy treatment planning.