Learned Cone-Beam CT Reconstruction Using Neural Ordinary Differential Equations

Thies M, Wagner F, Gu M, Folle L, Felsner L, Maier A (2022)

Publication Type: Conference contribution, Abstract of lecture

Publication year: 2022

Publisher: SPIE

Conference Proceedings Title: Proceedings Volume 12304, 7th International Conference on Image Formation in X-Ray Computed Tomography; 1230409 (2022)

Event location: Baltimore

DOI: 10.1117/12.2646442

Abstract

Learned iterative reconstruction algorithms for inverse problems offer the flexibility to combine analytical knowledge about the problem with modules learned from data. This way, they achieve high reconstruction performance while ensuring consistency with the measured data. In computed tomography, extending such approaches from 2D fan-beam to 3D cone-beam data is challenging due to the prohibitively high GPU memory that would be needed to train such models. This paper proposes to use neural ordinary differential equations to solve the reconstruction problem in a residual formulation via numerical integration. For training, there is no need to backpropagate through several unrolled network blocks nor through the internals of the solver. Instead, the gradients are obtained very memory-efficiently in the neural ODE setting allowing for training on a single consumer graphics card. The method is able to reduce the root mean squared error by over 30% compared to the best performing classical iterative reconstruction algorithm and produces high quality cone-beam reconstructions even in a sparse view scenario.

Authors with CRIS profile

Related research project(s)

Involved external institutions

King’s College London United Kingdom (GB)

How to cite

APA:

Thies, M., Wagner, F., Gu, M., Folle, L., Felsner, L., & Maier, A. (2022). Learned Cone-Beam CT Reconstruction Using Neural Ordinary Differential Equations. Paper presentation at 7th International Conference on Image Formation in X-Ray Computed Tomography, Baltimore, US.

MLA:

Thies, Mareike, et al. "Learned Cone-Beam CT Reconstruction Using Neural Ordinary Differential Equations." Presented at 7th International Conference on Image Formation in X-Ray Computed Tomography, Baltimore SPIE, 2022.

BibTeX: Download