A. Hauptmann, J. Adler, S. Arridge and O. Öktem, "Multi-Scale Learned Iterative Reconstruction," in IEEE Transactions on Computational Imaging, vol. 6, pp. 843-856, 2020, doi: 10.1109/TCI.2020.2990299
Multi-scale learned iterative reconstruction
|Author:||Hauptmann, Andreas1,2; Adler, Jonas3,4; Arridge, Simon2;|
1Research Unit of Mathematical Sciences; University of Oulu, Oulu, Finland
2Department of Computer Science; University College London, London, United Kingdom
3Elekta, Stockholm, Sweden
4KTH – Royal Institute of Technology, Stockolm, Sweden
5Department of Mathematics, KTH – Royal Institute of Technology, Stockholm, Sweden
|Online Access:||PDF Full Text (PDF, 3.9 MB)|
|Persistent link:|| http://urn.fi/urn:nbn:fi-fe2020052538879
Institute of Electrical and Electronics Engineers,
|Publish Date:|| 2020-05-25
Model-based learned iterative reconstruction methods have recently been shown to outperform classical reconstruction algorithms. Applicability of these methods to large scale inverse problems is however limited by the available memory for training and extensive training times, the latter due to computationally expensive forward models. As a possible solution to these restrictions we propose a multi-scale learned iterative reconstruction scheme that computes iterates on discretisations of increasing resolution. This procedure does not only reduce memory requirements, it also considerably speeds up reconstruction and training times, but most importantly is scalable to large scale inverse problems with non-trivial forward operators, such as those that arise in many 3D tomographic applications. In particular, we propose a hybrid network that combines the multi-scale iterative approach with a particularly expressive network architecture which in combination exhibits excellent scalability in 3D. Applicability of the algorithm is demonstrated for 3D cone beam computed tomography from real measurement data of an organic phantom. Additionally, we examine scalability and reconstruction quality in comparison to established learned reconstruction methods in two dimensions for low dose computed tomography on human phantoms.
IEEE transactions on computational imaging
|Pages:||843 - 856|
|Type of Publication:||
A1 Journal article – refereed
|Field of Science:||
113 Computer and information sciences
This work was partially supported by the Academy of Finland Project 312123 (Finnish Centre of Excellence in Inverse Modelling and Imaging, 2018–2025) and Project 334817, as well as British Heart Foundation grant NH/18/1/33511, EPSRC grant EP/M020533/1, and EPSRC-Wellcome grant WT101957. The authors would also like to acknowledge the Swedish Foundation for Strategic Research grants Low complexity reconstruction for medicine (AM13-0049) and 3D reconstruction with simulated image formation models (ID14-0055). Funding has also been provided by Elekta (Stockholm, Sweden).
|Academy of Finland Grant Number:||
312123 (Academy of Finland Funding decision)
334817 (Academy of Finland Funding decision)
© 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.