Traditional Machine Learning Techniques for Streak Artifact Reduction in Limited Angle Tomography

  • Yixing Huang
  • Yanye Lu
  • Oliver Taubmann
  • Guenter Lauritsch
  • Andreas Maier
Conference paper
Part of the Informatik aktuell book series (INFORMAT)

Zusammenfassung

In this work, the application of traditional machine learning techniques, in the form of regression models based on conventional, “hand-crafted” features, to streak reduction in limited angle tomography is investigated. Specifically, linear regression (LR), multi-layer perceptron (MLP), and reduced-error pruning tree (REPTree) are investigated. When choosing the mean-variation-median (MVM), Laplacian, and Hessian features, REPTree learns streak artifacts best and reaches the smallest root-mean-square error (RMSE) of 29HU for the Shepp-Logan phantom. Further experiments demonstrate that the MVM and Hessian features complement each other, whereas the Laplacian feature is redundant in the presence of MVM. Preliminary experiments on clinical data suggests that further investigation of clinical applications using REPTree may be worthwhile.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Literatur

  1. 1.
    Huang Y, Huang X, Taubmann O, et al. Restoration of missing data in limited angle tomography based on Helgason-Ludwig consistency conditions. Biomed Phys Eng Express. 2017;3(3).Google Scholar
  2. 2.
    Chen Z, Jin X, Li L, et al. A limited-angle CT reconstruction method based on anisotropic TV minimization. Biomed Phys Eng Express. 2013;58(7).Google Scholar
  3. 3.
    Huang Y, Taubmann O, Huang X, et al. A new scale space total variation algorithm for limited angle tomography. Proc CT-Meeting. 2016;3(3):149–152.Google Scholar
  4. 4.
    Würfl T, Ghesu F, Christlein V, et al. Deep learning computed tomography. Proc CT-Meeting. 2016;3(3):432–440.Google Scholar
  5. 5.
    Hammernik K, Würfl T, Pock T, et al. A deep learning architecture for limitedangle computed tomography reconstruction. Proc CT-Meeting. 2017;3(3):92–97.Google Scholar
  6. 6.
    Gu J, Ye J. Multi-scale wavelet domain residual learning for limited-angle CT reconstruction. Proc CT-Meeting. 2017;3(3):443–447.Google Scholar
  7. 7.
    Loh W. Classification and regression trees. Proc CT-Meeting. 2011;1(1):14–23.Google Scholar
  8. 8.
    Quinlan J. Simplifying decision trees. Proc CT-Meeting. 1987;27(3):221–234.Google Scholar
  9. 9.
    Frank E, Hall M, Witten I. The WEKA workbench: online appendix for data mining: practical machine learning tools and techniques.. vol. 27. Int J Man Mach Stud.; 2016.Google Scholar
  10. 10.
    Maier A, Berger M, Fischer P, et al. CONRAD: a software framework for conebeam imaging in radiology. Med Phys. 2013;40(11):111914.Google Scholar

Copyright information

© Springer-Verlag GmbH Deutschland 2018

Authors and Affiliations

  • Yixing Huang
    • 1
  • Yanye Lu
    • 1
  • Oliver Taubmann
    • 1
    • 2
  • Guenter Lauritsch
    • 3
  • Andreas Maier
    • 1
    • 2
  1. 1.Pattern Recognition LabFriedrich-Alexander-University Erlangen-NurembergErlangenDeutschland
  2. 2.Erlangen Graduate School in Advanced Optical Technologies (SAOT)ErlangenDeutschland
  3. 3.Siemens Healthcare GmbHForchheimDeutschland

Personalised recommendations