Deep learning based adaptive filtering for projection data noise reduction in x-ray computed tomography

2019 ◽  
Author(s):  
Tzu-Cheng Lee ◽  
Jian Zhou ◽  
Zhou Yu
Keyword(s):  
Computed Tomography ◽  
Deep Learning ◽  
Noise Reduction ◽  
Adaptive Filtering ◽  
Projection Data ◽  
Data Noise ◽  
X Ray ◽  
X Ray Computed

scholarly journals Deep learning based spectral extrapolation for dual‐source, dual‐energy x‐ray computed tomography

2020 ◽  
Vol 47 (9) ◽  
pp. 4150-4163
Author(s):  
Darin P. Clark ◽  
Fides R. Schwartz ◽  
Daniele Marin ◽  
Juan C. Ramirez‐Giraldo ◽  
Cristian T. Badea
Keyword(s):  
Computed Tomography ◽  
Deep Learning ◽  
Dual Energy ◽  
X Ray ◽  
Dual Source ◽  
X Ray Computed

scholarly journals A deep learning reconstruction framework for X-ray computed tomography with incomplete data

PLoS ONE ◽  
2019 ◽  
Vol 14 (11) ◽  
pp. e0224426 ◽  
Author(s):  
Jianbing Dong ◽  
Jian Fu ◽  
Zhao He
Keyword(s):  
Computed Tomography ◽  
Deep Learning ◽  
Incomplete Data ◽  
X Ray ◽  
X Ray Computed

scholarly journals Deep learning enabled ultra‐fast‐pitch acquisition in clinical X‐ray Computed Tomography

2021 ◽  
Author(s):  
Hao Gong ◽  
Liqiang Ren ◽  
Scott S. Hsieh ◽  
Cynthia H. McCollough ◽  
Lifeng Yu
Keyword(s):  
Computed Tomography ◽  
Deep Learning ◽  
X Ray ◽  
X Ray Computed

scholarly journals Level-set reconstruction algorithm for ultrafast limited-angle X-ray computed tomography of two-phase flows

2015 ◽  
Vol 373 (2043) ◽  
pp. 20140395 ◽  
Author(s):  
M. Bieberle ◽  
U. Hampel
Keyword(s):  
Computed Tomography ◽  
Level Set ◽  
Reconstruction Algorithm ◽  
Projection Data ◽  
Reconstruction Technique ◽  
Two Phase Flows ◽  
Two Phase ◽  
X Ray ◽  
X Ray Computed ◽  
Limited Angle

Tomographic image reconstruction is based on recovering an object distribution from its projections, which have been acquired from all angular views around the object. If the angular range is limited to less than 180° of parallel projections, typical reconstruction artefacts arise when using standard algorithms. To compensate for this, specialized algorithms using a priori information about the object need to be applied. The application behind this work is ultrafast limited-angle X-ray computed tomography of two-phase flows. Here, only a binary distribution of the two phases needs to be reconstructed, which reduces the complexity of the inverse problem. To solve it, a new reconstruction algorithm (LSR) based on the level-set method is proposed. It includes one force function term accounting for matching the projection data and one incorporating a curvature-dependent smoothing of the phase boundary. The algorithm has been validated using simulated as well as measured projections of known structures, and its performance has been compared to the algebraic reconstruction technique and a binary derivative of it. The validation as well as the application of the level-set reconstruction on a dynamic two-phase flow demonstrated its applicability and its advantages over other reconstruction algorithms.

Sign in / Sign up

Export Citation Format

Share Document