Sparse-View Image Reconstruction in Cone-Beam Computed Tomography with Variance-Reduced Stochastic Gradient Descent and Locally-Adaptive Proximal Operation

Abstract

There is a growing interest in image reconstruction from a small number of projections for computed tomography. Most of the available algorithms require a large number of iterations to reconstruct a high-quality image and they include parameters that need careful tuning. In this paper, we present a new algorithm that aims at reducing these problems. We formulate the reconstruction as an unconstrained optimization problem that consists of a measurement consistency term and a total variation regularization. The algorithm that we propose is based on the class of proximal gradient methods. Since the basic proximal gradient method is slow, we propose three modifications to improve its convergence speed. First, instead of proximal gradient iterations, we use a variance-reduced stochastic proximal gradient descent updates. Second, we apply the proximal operator with a locally adaptive regularization parameter; specifically, we partition the image into small blocks and denoise each block with a regularization parameter that depends on the probability of the presence of important image features in that block. Thirdly, at each iteration of the algorithm, we minimize the objective function over the subspace spanned by the current proximal gradient update and several previous update directions. The step size in the stochastic proximal gradient descent can be set equal to one and we suggest an easy method to find a small range that contains the acceptable values for the regularization parameter. Our experiments show that the proposed algorithm can recover a high-quality image from undersampled projections in a small number of iterations.

This is a preview of subscription content, access via your institution.

Algorithm 1:
Algorithm 2:
Fig. 1
Fig. 2
Algorithm 3:
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12

References

  1. 1.

    Berrington de Gonzlez, A., Mahesh, M., Kim, K., Bhargavan, M., Lewis, R., Mettler, F., et al. (2009). Projected cancer risks from computed tomographic scans performed in the United States in 2007. Archives of Internal Medicine, 169(22), 2071–2077.

    Article  Google Scholar 

  2. 2.

    Smith-Bindman, R., Lipson, J., Marcus, R., Kim, K., Mahesh, M., Gould, R., et al. (2009). Radiation dose associated with common computed tomography examinations and the associated lifetime attributable risk of cancer. Archives of Internal Medicine, 169(22), 2078–2086.

    Article  Google Scholar 

  3. 3.

    McCollough, C. H., Bruesewitz, M. R., & Kofler, J. M. (2006). CT dose reduction and dose management tools: Overview of available options. RadioGraphics, 26(2), 503–512.

    Article  Google Scholar 

  4. 4.

    McCollough, C. H., Primak, A. N., Braun, N., Kofler, J., Yu, L., & Christner, J. (2009). Strategies for reducing radiation dose in CT. Radiologic Clinics of North America, 47(1), 27–40.

    Article  Google Scholar 

  5. 5.

    Beister, M., Kolditz, D., & Kalender, W. A. (2012). Iterative reconstruction methods in x-ray CT. Physica Medica, 28(2), 94–108.

    Article  Google Scholar 

  6. 6.

    Feldkamp, L. A., Davis, L. C., & Kress, J. W. (1984). Practical cone-beam algorithm. Journal of the Optical Society of America A. Optics and Image Science, 1(6), 612–619.

    Article  Google Scholar 

  7. 7.

    Candes, E., Romberg, J., & Tao, T. (2006). Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information. IEEE Transactions on Information Theory, 52(2), 489–509.

    MathSciNet  Article  MATH  Google Scholar 

  8. 8.

    Sidky, E. Y., & Pan, X. (2008). Image reconstruction in circular cone-beam computed tomography by constrained, total-variation minimization. Physics in Medicine & Biology, 53(17), 4777–4807.

    Article  Google Scholar 

  9. 9.

    Choi, K., Wang, J., Zhu, L., Suh, T.-S., Boyd, S., & Xing, L. (2010). Compressed sensing based cone-beam computed tomography reconstruction with a first-order method. Medical Physics, 37(9), 5113–5125.

    Article  Google Scholar 

  10. 10.

    Park, J. C., Song, B., Kim, J. S., Park, S. H., Kim, H. K., Liu, Z., et al. (2012). Fast compressed sensing-based CBCT reconstruction using Barzilai-Borwein formulation for application to on-line IGRT. Medical Physics, 39(3), 1207–1217.

    Article  Google Scholar 

  11. 11.

    Richard, H. L. K., Van Metter, L., & Beutel, J. (2000). Handbook of medical imaging, physics and psychophysics (Vol. 1). Bellingham, WA: SPIE Publications.

    Google Scholar 

  12. 12.

    Macovski, A. (1983). Medical imaging systems. Upper Saddle River, NJ: Prentice Hall.

    Google Scholar 

  13. 13.

    Wang, J., Lu, H., Liang, Z., Eremina, D., Zhang, G., Wang, S., et al. (2008). An experimental study on the noise properties of x-ray CT sinogram data in Radon space. Physics in Medicine & Biology, 53(12), 3327.

    Article  Google Scholar 

  14. 14.

    Nuyts, J., De Man, B., Fessler, J. A., Zbijewski, W., & Beekman, F. J. (2013). Modelling the physics in the iterative reconstruction for transmission computed tomography. Physics in Medicine & Biology, 58(12), R63.

    Article  Google Scholar 

  15. 15.

    S. Zabic, Q. Wang, T. Morton, and K. M. Brown. A low dose simulation tool for ct systems with energy integrating detectors. Medical Physics, 40(3), 2013.

  16. 16.

    LaRivire, P. J. (2005). Penalized-likelihood sinogram smoothing for low-dose CT. Medical Physics, 32(6), 1676–1683.

    Article  Google Scholar 

  17. 17.

    Ramani, S., & Fessler, J. (2012). A splitting-based iterative algorithm for accelerated statistical x-ray ct reconstruction. IEEE Transactions on Medical Imaging, 31(3), 677–688.

    Article  Google Scholar 

  18. 18.

    De Man, B., & Basu, S. (2004). Distance-driven projection and backprojection in three dimensions. Physics in Medicine & Biology, 49(11), 2463–2475.

    Article  Google Scholar 

  19. 19.

    Long, Y., Fessler, J., & Balter, J. (2010). 3D forward and back-projection for x-ray CT using separable footprints. IEEE Transactions on Medical Imaging, 29(11), 1839–1850.

    Article  Google Scholar 

  20. 20.

    Ziegler, N. T. (2010). A, Khler T and P. R. Efficient projection and backprojection scheme for spherically symmetric basis functions in divergent beam geometry. Medical Physics, 33(12), 4653–4663.

    Article  Google Scholar 

  21. 21.

    Rudin, L. I., Osher, S., & Fatemi, E. (1992). Nonlinear total variation based noise removal algorithms. Physica D: Nonlinear Phenomena, 60(14), 259–268.

    MathSciNet  Article  MATH  Google Scholar 

  22. 22.

    Combettes, P. L. & Pesquet, J.-C. (2011). Proximal splitting methods in signal processing. In Fixed-point algorithms for inverse problems in science and engineering (pp. 185–212). New York: Springer.

  23. 23.

    Parikh, N., & Boyd, S. (2013). Proximal algorithms. Foundations and Trends in optimization, 1(3), 123–231.

    Google Scholar 

  24. 24.

    Chambolle, A. (2004). An algorithm for total variation minimization and applications. Journal of Mathematical Imaging and Vision, 20(1–2), 89–97.

    MathSciNet  Google Scholar 

  25. 25.

    Le Roux, N., Schmidt, M., & Bach, F. (2012). A stochastic gradient method with an exponential convergence rate for finite training sets. arXiv preprint arXiv:1202.6258.

  26. 26.

    Cevher, V., Becker, S., & Schmidt, M. (2014). Convex optimization for big data: Scalable, randomized, and parallel algorithms for big data analytics. IEEE Signal Processing Magazine, 31(5), 32–43.

    Article  Google Scholar 

  27. 27.

    Xiao, L., & Zhang, T. (2014). A proximal stochastic gradient method with progressive variance reduction. SIAM Journal on Optimization, 24(4), 2057–2075.

    MathSciNet  Article  MATH  Google Scholar 

  28. 28.

    Nesterov, Y. (2003). Introductory lectures on convex optimization: A basic course. Boston, MA: Kluwer.

    Google Scholar 

  29. 29.

    Beck, A., & Teboulle, M. (2009). A fast iterative shrinkage-thresholding algorithm for linear inverse problems. SIAM Journal on Imaging Sciences, 2(1), 183–202.

    MathSciNet  Article  MATH  Google Scholar 

  30. 30.

    Nesterov, Y. (1983). A method of solving a convex programming problem with convergence rate O (1/k2). Soviet Mathematics Doklady, 27(2), 372–376.

    MATH  Google Scholar 

  31. 31.

    Nemirovskii, A., & Yudin, D. (1983). Problem complexity and method efficiency in optimization. Chichester: Wiley.

    Google Scholar 

  32. 32.

    Johnson, R., & Zhang, T. (2013). Accelerating stochastic gradient descent using predictive variance reduction. In C. Burges, L. Bottou, M. Welling, Z. Ghahramani, & K. Weinberger (Eds.), Advances in neural information processing systems (pp. 315–323). Red Hook: Curran Associates Inc.

    Google Scholar 

  33. 33.

    Defazio, A., Bach, F., & Lacoste-Julien, S. (2014). Saga: A fast incremental gradient method with support for non-strongly convex composite objectives. In Advances in neural information processing systems (pp. 1646–1654).

  34. 34.

    Mahdavi, M. & Jin, R. (2013). MixedGrad: An O(1/T) convergence rate algorithm for stochastic smooth optimization. arXiv preprint arXiv:1307.7192.

  35. 35.

    Combettes, P. L., & Wajs, V. R. (2005). Signal recovery by proximal forward-backward splitting. Multiscale Modeling & Simulation, 4(4), 1168–1200.

    MathSciNet  Article  MATH  Google Scholar 

  36. 36.

    Bioucas-Dias, J., & Figueiredo, M. A. T. (2007). A new twist: Two-step iterative shrinkage/thresholding algorithms for image restoration. IEEE Transactions on Image Processing, 16(12), 2992–3004.

    MathSciNet  Article  Google Scholar 

  37. 37.

    Nesterov, Y. (2007). Gradient methods for minimizing composite objective function. Technical Report CCIT 559, Universite Catholique de Louvain.

  38. 38.

    Elad, M., Matalon, B., & Zibulevsky, M. (2007). Coordinate and subspace optimization methods for linear least squares with non-quadratic regularization. Applied and Computational Harmonic Analysis, 23(3), 346–367.

    MathSciNet  Article  MATH  Google Scholar 

  39. 39.

    Louchet, C., & Moisan, L. (2011). Total variation as a local filter. SIAM Journal on Imaging Sciences, 4(2), 651–694.

    MathSciNet  Article  MATH  Google Scholar 

  40. 40.

    Strong, D., & Chan, T. (2003). Edge-preserving and scale-dependent properties of total variation regularization. Inverse Problems, 19(6), 165–187.

    MathSciNet  Article  MATH  Google Scholar 

  41. 41.

    Du, L. Y., Umoh, J., Nikolov, H. N., Pollmann, S. I., Lee, T. Y., & Holdsworth, D. W. (2007). A quality assurance phantom for the performance evaluation of volumetric micro-CT systems. Physics in Medicine & Biology, 52(23), 7087–7108.

    Article  Google Scholar 

  42. 42.

    Beck, A., & Teboulle, M. (2009). Fast gradient-based algorithms for constrained total variation image denoising and deblurring problems. Image Processing, IEEE Transactions on, 18(11), 2419–2434.

    MathSciNet  Article  Google Scholar 

  43. 43.

    Wang, Z., Bovik, A., Sheikh, H., & Simoncelli, E. (2004). Image quality assessment: from error visibility to structural similarity. IEEE Transactions on Image Processing, 13(4), 600–612.

    Article  Google Scholar 

  44. 44.

    Pluim, J., Maintz, J., & Viergever, M. (2003). Mutual-information-based registration of medical images: a survey. IEEE Transactions on Medical Imaging, 22(8), 986–1004.

    Article  MATH  Google Scholar 

  45. 45.

    Bian, J., Siewerdsen, J. H., Han, X., Sidky, E. Y., Prince, J. L., Pelizzari, C. A., et al. (2010). Evaluation of sparse-view reconstruction from at-panel-detector cone-beam CT. Physics in Medicine & Biology, 55(22), 6575.

    Article  Google Scholar 

  46. 46.

    Buhr, E., Günther-Kohfahl, S., & Neitzel, U. (2003). Simple method for modulation transfer function determination of digital imaging detectors from edge images. In Medical imaging 2003 (pp. 877–884). International Society for Optics and Photonics.

  47. 47.

    Wang, J., Guan, H., & Solberg, T. (2011). Inverse determination of the penalty parameter in penalized weighted least-squares algorithm for noise reduction of low-dose cbct. Medical Physics, 38(7), 4066–4072.

    Article  Google Scholar 

  48. 48.

    Eldar, Y. C. (2009). Generalized sure for exponential families: Applications to regularization. IEEE Transactions on Signal Processing, 57(2), 471–481.

    MathSciNet  Article  Google Scholar 

  49. 49.

    Riddell, C., & Trousset, Y. (2006). Rectification for cone-beam projection and backprojection. IEEE Transactions on Medical Imaging, 25(7), 950–962.

    Article  Google Scholar 

  50. 50.

    Basu, S., & Bresler, Y. (2002). o(n3log(n))backprojection algorithm for the 3-d radon transform. IEEE Transactions on Medical Imaging, 21(2), 76–88.

    Article  Google Scholar 

  51. 51.

    Zibulevsky, M., & Elad, M. (2010). L1-l2 optimization in signal and image processing. IEEE Signal Processing Magazine, 27(3), 76–88.

    Article  Google Scholar 

  52. 52.

    Dong, Y., & Hintermller, M. (2009). Multi-scale total variation with automated regularization parameter selection for color image restoration. In X.-C. Tai, K. Mrken, M. Lysaker, & K.-A. Lie (Eds.), Scale space and variational methods in computer vision (Vol. 5567, pp. 271–281)., Lecture Notes in Computer Science Berlin Heidelberg: Springer.

    Google Scholar 

  53. 53.

    Abbey, C. K., Sohl-Dickstein, J. N., Olshausen, B. A., Eckstein, M. P., & Boone, J. M. (2009). Higher-order scene statistics of breast images

  54. 54.

    Strong, D. M., Aujol, J.-F., & Chan, T. F. (2006). Scale recognition, regularization parameter selection, and meyer’s g norm in total variation regularization. Multiscale Modeling & Simulation, 5(1), 273–303.

    MathSciNet  Article  MATH  Google Scholar 

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Davood Karimi.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Karimi, D., Ward, R.K. Sparse-View Image Reconstruction in Cone-Beam Computed Tomography with Variance-Reduced Stochastic Gradient Descent and Locally-Adaptive Proximal Operation. J. Med. Biol. Eng. 37, 420–440 (2017). https://doi.org/10.1007/s40846-017-0231-7

Download citation

Keywords

  • Iterative reconstruction
  • Cone-beam CT
  • Compressive sensing
  • Projected gradient
  • Stochastic
  • Gradient descent