TY - GEN
T1 - JSR-Net
T2 - 44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019
AU - Zhang, Haimiao
AU - Dong, Bin
AU - Liu, Baodong
N1 - Funding Information:
∗The work of this author is funded by China Postdoctoral Science Foundation under grant 2018M641056. †Corresponding author. The research of this author is supported by NSFC grant 11831002. Email: dongbin@math.pku.edu.cn ‡Corresponding author. This work is supported by the National Key Scientific Instrument and Equipment Development Project 2017YFF0107200.
Publisher Copyright:
© 2019 IEEE.
PY - 2019/5
Y1 - 2019/5
N2 - CT image reconstruction from incomplete data, such as sparse views and limited angle reconstruction, is an important and challenging problem in medical imaging. This work proposes a new deep convolutional neural network (CNN), called JSR-Net, that jointly reconstructs CT images and their associated Radon domain projections. JSR-Net combines the traditional model based approach with deep architecture design of deep learning. A hybrid loss function is adopted to improve the performance of the JSR-Net making it more effective in protecting important image structures. Numerical experiments demonstrate that JSR-Net outperforms some latest model based reconstruction methods, as well as a recently proposed deep model.
AB - CT image reconstruction from incomplete data, such as sparse views and limited angle reconstruction, is an important and challenging problem in medical imaging. This work proposes a new deep convolutional neural network (CNN), called JSR-Net, that jointly reconstructs CT images and their associated Radon domain projections. JSR-Net combines the traditional model based approach with deep architecture design of deep learning. A hybrid loss function is adopted to improve the performance of the JSR-Net making it more effective in protecting important image structures. Numerical experiments demonstrate that JSR-Net outperforms some latest model based reconstruction methods, as well as a recently proposed deep model.
KW - Convolutional neural networks
KW - Deep learning
KW - Joint spatial-Radon domain reconstruction
KW - Limited angle CT
KW - Sparse-view CT
UR - http://www.scopus.com/inward/record.url?scp=85069005383&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85069005383&partnerID=8YFLogxK
U2 - 10.1109/ICASSP.2019.8682178
DO - 10.1109/ICASSP.2019.8682178
M3 - Conference contribution
AN - SCOPUS:85069005383
T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
SP - 3657
EP - 3661
BT - 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 12 May 2019 through 17 May 2019
ER -