A deep learning reconstruction framework for X-ray computed tomography with incomplete data

被引:54
作者
Dong, Jianbing [1 ]
Fu, Jian [1 ,2 ,3 ]
He, Zhao [1 ]
机构
[1] Beijing Univ Aeronaut & Astronaut, Res Ctr Digital Radiat Imaging & Biomed Imaging, Beijing, Peoples R China
[2] Beijing Univ Aeronaut & Astronaut, Sch Mech Engn & Automat, Beijing, Peoples R China
[3] Beijing Univ Aeronaut & Astronaut, Jiangxi Res Inst, Nanchang, Jiangxi, Peoples R China
基金
中国国家自然科学基金;
关键词
CONVOLUTIONAL NEURAL-NETWORK; IMAGE-RECONSTRUCTION; ART;
D O I
10.1371/journal.pone.0224426
中图分类号
O [数理科学和化学]; P [天文学、地球科学]; Q [生物科学]; N [自然科学总论];
学科分类号
07 ; 0710 ; 09 ;
摘要
As a powerful imaging tool, X-ray computed tomography (CT) allows us to investigate the inner structures of specimens in a quantitative and nondestructive way. Limited by the implementation conditions, CT with incomplete projections happens quite often. Conventional reconstruction algorithms are not easy to deal with incomplete data. They are usually involved with complicated parameter selection operations, also sensitive to noise and timeconsuming. In this paper, we reported a deep learning reconstruction framework for incomplete data CT. It is the tight coupling of the deep learning U-net and CT reconstruction algorithm in the domain of the projection sinograms. The U-net estimated results are not the artifacts caused by the incomplete data, but the complete projection sinograms. After training, this framework is determined and can reconstruct the final high quality CT image from a given incomplete projection sinogram. Taking the sparse-view and limited-angle CT as examples, this framework has been validated and demonstrated with synthetic and experimental data sets. Embedded with CT reconstruction, this framework naturally encapsulates the physical imaging model of CT systems and is easy to be extended to deal with other challenges. This work is helpful to push the application of the state-of-the-art deep learning techniques in the field of CT.
引用
收藏
页数:17
相关论文
共 33 条
[1]   SIMULTANEOUS ALGEBRAIC RECONSTRUCTION TECHNIQUE (SART) - A SUPERIOR IMPLEMENTATION OF THE ART ALGORITHM [J].
ANDERSEN, AH ;
KAK, AC .
ULTRASONIC IMAGING, 1984, 6 (01) :81-94
[2]  
[Anonymous], 2015, ARXIV PREPRINT ARXIV
[3]  
[Anonymous], ARXIV170301382V1
[4]  
[Anonymous], ARXIV161106391V2
[5]  
[Anonymous], ICLR 2016
[6]   LEARNING LONG-TERM DEPENDENCIES WITH GRADIENT DESCENT IS DIFFICULT [J].
BENGIO, Y ;
SIMARD, P ;
FRASCONI, P .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 1994, 5 (02) :157-166
[7]  
Burger HC, 2012, PROC CVPR IEEE, P2392, DOI 10.1109/CVPR.2012.6247952
[8]   Prior image constrained compressed sensing (PICCS): A method to accurately reconstruct dynamic CT images from highly undersampled projection data sets [J].
Chen, Guang-Hong ;
Tang, Jie ;
Leng, Shuai .
MEDICAL PHYSICS, 2008, 35 (02) :660-663
[10]   Compression Artifacts Reduction by a Deep Convolutional Network [J].
Dong, Chao ;
Deng, Yubin ;
Loy, Chen Change ;
Tang, Xiaoou .
2015 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV), 2015, :576-584