Regularized CT reconstruction on unstructured grid

Yun Chen; Yao Lu; Xiangyuan Ma; Yuesheng Xu

doi:10.1117/12.2217381

1 April 2016 Regularized CT reconstruction on unstructured grid

Yun Chen, Yao Lu, Xiangyuan Ma, Yuesheng Xu

Proceedings Volume 9783, Medical Imaging 2016: Physics of Medical Imaging; 97834G (2016) https://doi.org/10.1117/12.2217381
Event: SPIE Medical Imaging, 2016, San Diego, California, United States

Abstract

Computed tomography (CT) is an ill-posed problem. Reconstruction on unstructured grid reduces the computational cost and alleviates the ill-posedness by decreasing the dimension of the solution space. However, there was no systematic study on edge-preserving regularization methods for CT reconstruction on unstructured grid. In this work, we propose a novel regularization method for CT reconstruction on unstructured grid, such as triangular or tetrahedral meshes generated from the initial images reconstructed via analysis reconstruction method (e.g., filtered back-projection). The proposed regularization method is modeled as a three-term optimization problem, containing a weighted least square fidelity term motivated by the simultaneous algebraic reconstruction technique (SART). The related cost function contains two non-differentiable terms, which bring difficulty to the development of the fast solver. A fixed-point proximity algorithm with SART is developed for solving the related optimization problem, and accelerating the convergence. Finally, we compare the regularized CT reconstruction method to SART with different regularization methods. Numerical experiments demonstrated that the proposed regularization method on unstructured grid is effective to suppress noise and preserve edge features.

Citation Download Citation

Yun Chen, Yao Lu, Xiangyuan Ma, and Yuesheng Xu "Regularized CT reconstruction on unstructured grid", Proc. SPIE 9783, Medical Imaging 2016: Physics of Medical Imaging, 97834G (1 April 2016); https://doi.org/10.1117/12.2217381

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available