Literature DB >> 33259647

Improving CBCT quality to CT level using deep learning with generative adversarial network.

Yang Zhang1,2, Ning Yue1, Min-Ying Su2, Bo Liu1, Yi Ding3, Yongkang Zhou4, Hao Wang5, Yu Kuang6, Ke Nie1.   

Abstract

PURPOSE: To improve image quality and computed tomography (CT) number accuracy of daily cone beam CT (CBCT) through a deep learning methodology with generative adversarial network.
METHODS: One hundred and fifty paired pelvic CT and CBCT scans were used for model training and validation. An unsupervised deep learning method, 2.5D pixel-to-pixel generative adversarial network (GAN) model with feature mapping was proposed. A total of 12 000 slice pairs of CT and CBCT were used for model training, while ten-fold cross validation was applied to verify model robustness. Paired CT-CBCT scans from an additional 15 pelvic patients and 10 head-and-neck (HN) patients with CBCT images collected at a different machine were used for independent testing purpose. Besides the proposed method above, other network architectures were also tested as: 2D vs 2.5D; GAN model with vs without feature mapping; GAN model with vs without additional perceptual loss; and previously reported models as U-net and cycleGAN with or without identity loss. Image quality of deep-learning generated synthetic CT (sCT) images was quantitatively compared against the reference CT (rCT) image using mean absolute error (MAE) of Hounsfield units (HU) and peak signal-to-noise ratio (PSNR). The dosimetric calculation accuracy was further evaluated with both photon and proton beams.
RESULTS: The deep-learning generated sCTs showed improved image quality with reduced artifact distortion and improved soft tissue contrast. The proposed algorithm of 2.5 Pix2pix GAN with feature matching (FM) was shown to be the best model among all tested methods producing the highest PSNR and the lowest MAE to rCT. The dose distribution demonstrated a high accuracy in the scope of photon-based planning, yet more work is needed for proton-based treatment. Once the model was trained, it took 11-12 ms to process one slice, and could generate a 3D volume of dCBCT (80 slices) in less than a second using a NVIDIA GeForce GTX Titan X GPU (12 GB, Maxwell architecture).
CONCLUSION: The proposed deep learning algorithm is promising to improve CBCT image quality in an efficient way, thus has a potential to support online CBCT-based adaptive radiotherapy.
© 2020 American Association of Physicists in Medicine.

Entities:  

Keywords:  CBCT; GAN; adaptive radiotherapy; deep learning

Mesh:

Year:  2021        PMID: 33259647      PMCID: PMC8166936          DOI: 10.1002/mp.14624

Source DB:  PubMed          Journal:  Med Phys        ISSN: 0094-2405            Impact factor:   4.071


  26 in total

1.  Improved scatter correction using adaptive scatter kernel superposition.

Authors:  M Sun; J M Star-Lack
Journal:  Phys Med Biol       Date:  2010-10-28       Impact factor: 3.609

2.  Proton dose calculation on scatter-corrected CBCT image: Feasibility study for adaptive proton therapy.

Authors:  Yang-Kyun Park; Gregory C Sharp; Justin Phillips; Brian A Winey
Journal:  Med Phys       Date:  2015-08       Impact factor: 4.071

3.  Efficient Monte Carlo based scatter artifact reduction in cone-beam micro-CT.

Authors:  Wojciech Zbijewski; Freek J Beekman
Journal:  IEEE Trans Med Imaging       Date:  2006-07       Impact factor: 10.048

4.  Scatter correction of cone-beam CT using a deep residual convolution neural network (DRCNN).

Authors:  Yangkang Jiang; Chunlin Yang; Pengfei Yang; Xi Hu; Chen Luo; Yi Xue; Lei Xu; Xiuhua Hu; Luhan Zhang; Jing Wang; Ke Sheng; Tianye Niu
Journal:  Phys Med Biol       Date:  2019-07-11       Impact factor: 3.609

5.  ScatterNet: A convolutional neural network for cone-beam CT intensity correction.

Authors:  David C Hansen; Guillaume Landry; Florian Kamp; Minglun Li; Claus Belka; Katia Parodi; Christopher Kurz
Journal:  Med Phys       Date:  2018-10-08       Impact factor: 4.071

6.  Male pelvic multi-organ segmentation aided by CBCT-based synthetic MRI.

Authors:  Yang Lei; Tonghe Wang; Sibo Tian; Xue Dong; Ashesh B Jani; David Schuster; Walter J Curran; Pretesh Patel; Tian Liu; Xiaofeng Yang
Journal:  Phys Med Biol       Date:  2020-02-04       Impact factor: 3.609

7.  Synthetic CT generation from CBCT images via deep learning.

Authors:  Liyuan Chen; Xiao Liang; Chenyang Shen; Steve Jiang; Jing Wang
Journal:  Med Phys       Date:  2020-01-13       Impact factor: 4.071

8.  A GPU tool for efficient, accurate, and realistic simulation of cone beam CT projections.

Authors:  Xun Jia; Hao Yan; Laura Cervino; Michael Folkerts; Steve B Jiang
Journal:  Med Phys       Date:  2012-12       Impact factor: 4.071

9.  Paired cycle-GAN-based image correction for quantitative cone-beam computed tomography.

Authors:  Joseph Harms; Yang Lei; Tonghe Wang; Rongxiao Zhang; Jun Zhou; Xiangyang Tang; Walter J Curran; Tian Liu; Xiaofeng Yang
Journal:  Med Phys       Date:  2019-07-17       Impact factor: 4.071

10.  Projection-domain scatter correction for cone beam computed tomography using a residual convolutional neural network.

Authors:  Yusuke Nomura; Qiong Xu; Hiroki Shirato; Shinichi Shimizu; Lei Xing
Journal:  Med Phys       Date:  2019-06-05       Impact factor: 4.071

View more
  6 in total

1.  A Comparison Study Between CNN-Based Deformed Planning CT and CycleGAN-Based Synthetic CT Methods for Improving iCBCT Image Quality.

Authors:  Bo Yang; Yankui Chang; Yongguang Liang; Zhiqun Wang; Xi Pei; Xie George Xu; Jie Qiu
Journal:  Front Oncol       Date:  2022-05-30       Impact factor: 5.738

2.  Simple Code Implementation for Deep Learning-Based Segmentation to Evaluate Central Serous Chorioretinopathy in Fundus Photography.

Authors:  Tae Keun Yoo; Bo Yi Kim; Hyun Kyo Jeong; Hong Kyu Kim; Donghyun Yang; Ik Hee Ryu
Journal:  Transl Vis Sci Technol       Date:  2022-02-01       Impact factor: 3.283

3.  Generating synthesized computed tomography from CBCT using a conditional generative adversarial network for head and neck cancer patients.

Authors:  Yun Zhang; Sheng-Gou Ding; Xiao-Chang Gong; Xing-Xing Yuan; Jia-Fan Lin; Qi Chen; Jin-Gao Li
Journal:  Technol Cancer Res Treat       Date:  2022 Jan-Dec

4.  A more effective CT synthesizer using transformers for cone-beam CT-guided adaptive radiotherapy.

Authors:  Xinyuan Chen; Yuxiang Liu; Bining Yang; Ji Zhu; Siqi Yuan; Xuejie Xie; Yueping Liu; Jianrong Dai; Kuo Men
Journal:  Front Oncol       Date:  2022-08-25       Impact factor: 5.738

Review 5.  Deep learning methods for enhancing cone-beam CT image quality toward adaptive radiation therapy: A systematic review.

Authors:  Branimir Rusanov; Ghulam Mubashar Hassan; Mark Reynolds; Mahsheed Sabet; Jake Kendrick; Pejman Rowshanfarzad; Martin Ebert
Journal:  Med Phys       Date:  2022-07-18       Impact factor: 4.506

6.  Clinical suitability of deep learning based synthetic CTs for adaptive proton therapy of lung cancer.

Authors:  Adrian Thummerer; Carmen Seller Oria; Paolo Zaffino; Arturs Meijers; Gabriel Guterres Marmitt; Robin Wijsman; Joao Seco; Johannes Albertus Langendijk; Antje-Christin Knopf; Maria Francesca Spadea; Stefan Both
Journal:  Med Phys       Date:  2021-11-16       Impact factor: 4.506

  6 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.