Reconstructing invariances of CT image denoising networks using invertible neural networks

Elias Eulig; Björn Ommer; Marc Kachelrieß

doi:10.1117/12.2647170

17 October 2022 Reconstructing invariances of CT image denoising networks using invertible neural networks

Elias Eulig, Björn Ommer, Marc Kachelrieß

Author Affiliations +

Proceedings Volume 12304, 7th International Conference on Image Formation in X-Ray Computed Tomography; 123040S (2022) https://doi.org/10.1117/12.2647170
Event: Seventh International Conference on Image Formation in X-Ray Computed Tomography (ICIFXCT 2022), 2022, Baltimore, United States

Abstract

Long lasting efforts have been made to reduce radiation dose and thus the potential radiation risk to the patient for CT acquisitions without severe deterioration of image quality. To this end, different reconstruction and noise reduction algorithms have been developed, many of which are based on iterative reconstruction techniques, incorporating prior knowledge in the image domain. Recently, deep learning-based methods have shown impressive performance, outperforming many of the previously proposed CT denoising approaches both visually and quantitatively. However, with most neural networks being black boxes they remain notoriously difficult to interpret and concerns about the robustness and safety of such denoising methods have been raised. In this work we want to lay the fundamentals for a post-hoc interpretation of existing CT denoising networks by reconstructing their invariances.

Citation Download Citation

Elias Eulig, Björn Ommer, and Marc Kachelrieß "Reconstructing invariances of CT image denoising networks using invertible neural networks", Proc. SPIE 12304, 7th International Conference on Image Formation in X-Ray Computed Tomography, 123040S (17 October 2022); https://doi.org/10.1117/12.2647170

ACCESS THE FULL ARTICLE

PROCEEDINGS
5 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Denoising

Computed tomography

Neural networks

Image denoising

Reconstruction algorithms

Image quality

RELATED CONTENT

A task informed model training method for deep neural network...
Proceedings of SPIE (April 04 2022)

Tunable neural networks for multi material image formation from spectral...
Proceedings of SPIE (October 17 2022)

Deep neural networks based denoising models for CT imaging and...
Proceedings of SPIE (February 15 2021)

2.5D dictionary learning based computed tomography reconstruction
Proceedings of SPIE (May 12 2016)

Multidimensional noise reduction in C arm cone beam CT via...
Proceedings of SPIE (March 01 2019)

Generative adversarial networks based regularized image reconstruction for PET
Proceedings of SPIE (May 28 2019)

Subscribe to Digital Library

Receive Erratum Email Alert

Keywords/Phrases

Search In:

Publication Years