Dictionary Learning based Low Dose Helical CT Reconstruction With Longitudinal TV Constraint

Yongyi Shi, Hengyong Yu, Yanbo Zhang, Rui Liu, Mannudeep Kalra, Ge Wang, Xuanqin Mou


Published in:Fully3D 2017 Proceedings


low dose HCT, longitudinal constraint, dictionary learning, distance driven, GPU implementation
Multi-slice helical Computed Tomography (HCT) has been widely applied in clinical applications. Due to the potential radiation risk, it has attracted an increasing attention to reduce radiation dose while maintaining the diagnostic performance. Inspired by the longitudinal sampling inconsistencies of helical CT scanning, in this paper, we develop a statistical iterative reconstruction algorithm based on three-dimensional dictionary learning to improve image quality for low-dose HCT. The longitudinal Total Variation (TV) is added to change the image noise distribution. The classical distance-driven projection and back-projection models are employed to avoid artifact-inducing. To enhance the computational performance, Graphics Processing Unit (GPU) implementation, Order Subset technology and Nesterov’s acceleration strategy are employed in our iterative reconstruction codes to accelerate the optimization. The Contrast Noise Ratio (CNR) index of reconstructed images and the subjective evaluation of medical practitioners all verify the superiority of our proposed algorithm.
Yongyi Shi
Institute of Image Processing and Pattern Recognition, Xi'an Jiaotong University, China
Hengyong Yu
University of Massachusetts Lowell, USA
Yanbo Zhang
University of Massachusetts Lowell, USA
Rui Liu
University of Massachusetts Lowell, USA
Mannudeep Kalra
Massachusetts General Hospital, USA
Ge Wang
Biomedical Imaging Center, Rensselaer Polytechnic Institute, USA
Xuanqin Mou
Institute of Image Processing and Pattern Recognition, Xi'an Jiaotong University, China
  1. Kachelrieß M, Schaller S, Kalender W A. Advanced single-slice rebinning in cone-beam spiral CT[J]. Medical Physics, 2000, 27(4):754-772.
  2. Flohr T, Stierstorfer K, Bruder H, et al. Image reconstruction and image quality evaluation for a 16‐slice CT scanner[J]. Medical Physics, 2003, 30(5):832-845.
  3. Stierstorfer K, Rauscher A, Boese J, et al. Weighted FBP--a simple approximate 3D FBP algorithm for multislice spiral CT with good dose usage for arbitrary pitch.[J]. Physics in Medicine & Biology, 2004, 49(11):2209-2218.
  4. Thibault J B, Sauer K D, Bouman C A, et al. A three-dimensional statistical approach to improved image quality for multislice helical CT.[J]. Medical Physics, 2007, 34(11):4526.
  5. Sidky E, Pan X. Image reconstruction in circular cone-beam computed tomography by constrained, total-variation minimization.[J]. Physics in Medicine & Biology, 2008, 53(17):4777-4807.
  6. Xu Q, Yu H, Mou X, et al. Low-dose X-ray CT reconstruction via dictionary learning.[J]. IEEE Transactions on Medical Imaging, 2012, 31(9):1682-1697.
  7. De M B, Basu S. Distance-driven projection and backprojection in three dimensions.[J]. Physics in Medicine & Biology, 2004, 49(11):2463-2475.
  8. D. Kim, S. Ramani, and J. A. Fessler, “Accelerating X-ray CT ordered subsets image reconstruction with Nesterov’s first-order methods,” in Proc. Intl. Mtg. on Fully 3D Image Recon. in Rad. and Nuc. Med, pp. 22–5, 2013.
  9. http://www.aapm.org/GrandChallenge/LowDoseCT/.
  10. Flohr T G, Stierstorfer K, Ulzheimer S, et al. Image reconstruction and image quality evaluation for a 64‐slice CT scanner with z‐flying focal spot[J]. Medical Physics, 2005, 32(8):2536-254