|
Title
|
|
|
|
Ring artifact reduction in sinogram space using deep learning
| |
|
Author
|
|
|
|
| |
|
Abstract
|
|
|
|
| Ring artifacts are a type of reconstruction artifact that is common in X-Ray CT. Recently, methods based on deep learning have been proposed to reduce ring artifacts in reconstructed images. These methods are dependent on the choice of reconstruction algorithm and often rely on a polar coordinate transformation. Methods that directly operate in sinogram space do not feature this dependency, do not require a coordinate transformation while also operating in the space where ring artifacts originate. In this paper, we propose a deep neural network with a custom loss function that operates exclusively in sinogram space for ring artifact reduction. Results on real and simulated data show that our method has similar or better performance compared to other ring artifact reduction techniques that also operate exclusively in sinogram space. |
| |
|
Language
|
|
|
|
English
| |
|
Source (book)
|
|
|
|
The 6th International Conference on Image Formation in X-Ray Computed Tomography, 3-7 August, 2020, Regensburg, Germany
| |
|
Publication
|
|
|
|
2020
| |
|
Volume/pages
|
|
|
|
(2020)
, p. 486-489
| |
|
Full text (open access)
|
|
|
|
| |
|