Simultaneous Reconstruction and Segmentation of CT Scans with Shadowed Data
We propose a variational approach for simultaneous reconstruction and multiclass segmentation of X-ray CT images, with limited field of view and missing data. We propose a simple energy minimisation approach, loosely based on a Bayesian rationale. The resulting non convex problem is solved by alternating reconstruction steps using an iterated relaxed proximal gradient, and a proximal approach for the segmentation. Preliminary results on synthetic data demonstrate the potential of the approach for synchrotron imaging applications.
KeywordsFilter Back Projection Algebraic Reconstruction Technique Discrete Image Inverse Radon Filter Back Projection Reconstruction
F. Lauze acknowledges funding from the Innovation Fund Denmark and Mærsk Oil and Gas A/S, for the P\(^3\) Project. E. Plenge acknowledges funding from EUs 7FP under the Marie Skodowska-Curie grant (agreement no 600207), and from the Danish Council for Independent Research (grant ID DFF5054-00218).
- 4.Buzug, T.: Computed Tomography: From Photon Statistics to Modern Cone Beam CT. Springer, Heidelberg (2008)Google Scholar
- 5.Frickel, J., Quinto, E.T.: Characterization and reduction of artifacts in limited angle tomography. Inverse Probl. 29(12) (2013)Google Scholar
- 9.Romanov, M., Dahl, A.B., Dong, Y., Hansen, P.C.: Simultaneous tomographic reconstruction and segmentation with class priors. Inverse Probl. Sci. Eng. 24(8) (2016)Google Scholar
- 10.van de Sompel, D., Brady, M.: Simultaneous reconstruction and segmentation algorithm for positron emission tomography and transmission tomography. In: Proceedings of the 2008 International Symposium on Biomedical Imaging (2008)Google Scholar
- 11.Storah, M., Weinmann, A., Frickel, J., Unser, M.: Joint image reconstruction and segmentation using the Potts model. Inverse Probl. 2(32) (2015)Google Scholar