A Flexible Meta Learning Model for Image Registration

Frederic Kanter, Jan Lellmann
Proceedings of The 5th International Conference on Medical Imaging with Deep Learning, PMLR 172:638-652, 2022.

Abstract

We propose a trainable architecture for affine image registration to produce robust starting points for conventional image registration methods. Learning-based methods for image registration often require networks with many parameters and heavily engineered cost functions and thus are complex and computationally expensive. Despite their success in recent years, these methods often lack the accuracy of classical iterative image registration and struggle with large deformations. On the other hand, iterative methods depend on good initial estimates and tuned hyperparameters. We tackle this problem by combining effective shallow networks and classical optimization algorithms using strategies from the field of meta-learning. The architecture presented in this work incorporates only first-order gradient information of the given registration problems, making it highly flexible and particularly well-suited as an initialization step for classical image registration.

Cite this Paper


BibTeX
@InProceedings{pmlr-v172-kanter22a, title = {A Flexible Meta Learning Model for Image Registration}, author = {Kanter, Frederic and Lellmann, Jan}, booktitle = {Proceedings of The 5th International Conference on Medical Imaging with Deep Learning}, pages = {638--652}, year = {2022}, editor = {Konukoglu, Ender and Menze, Bjoern and Venkataraman, Archana and Baumgartner, Christian and Dou, Qi and Albarqouni, Shadi}, volume = {172}, series = {Proceedings of Machine Learning Research}, month = {06--08 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v172/kanter22a/kanter22a.pdf}, url = {https://proceedings.mlr.press/v172/kanter22a.html}, abstract = {We propose a trainable architecture for affine image registration to produce robust starting points for conventional image registration methods. Learning-based methods for image registration often require networks with many parameters and heavily engineered cost functions and thus are complex and computationally expensive. Despite their success in recent years, these methods often lack the accuracy of classical iterative image registration and struggle with large deformations. On the other hand, iterative methods depend on good initial estimates and tuned hyperparameters. We tackle this problem by combining effective shallow networks and classical optimization algorithms using strategies from the field of meta-learning. The architecture presented in this work incorporates only first-order gradient information of the given registration problems, making it highly flexible and particularly well-suited as an initialization step for classical image registration.} }
Endnote
%0 Conference Paper %T A Flexible Meta Learning Model for Image Registration %A Frederic Kanter %A Jan Lellmann %B Proceedings of The 5th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2022 %E Ender Konukoglu %E Bjoern Menze %E Archana Venkataraman %E Christian Baumgartner %E Qi Dou %E Shadi Albarqouni %F pmlr-v172-kanter22a %I PMLR %P 638--652 %U https://proceedings.mlr.press/v172/kanter22a.html %V 172 %X We propose a trainable architecture for affine image registration to produce robust starting points for conventional image registration methods. Learning-based methods for image registration often require networks with many parameters and heavily engineered cost functions and thus are complex and computationally expensive. Despite their success in recent years, these methods often lack the accuracy of classical iterative image registration and struggle with large deformations. On the other hand, iterative methods depend on good initial estimates and tuned hyperparameters. We tackle this problem by combining effective shallow networks and classical optimization algorithms using strategies from the field of meta-learning. The architecture presented in this work incorporates only first-order gradient information of the given registration problems, making it highly flexible and particularly well-suited as an initialization step for classical image registration.
APA
Kanter, F. & Lellmann, J.. (2022). A Flexible Meta Learning Model for Image Registration. Proceedings of The 5th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 172:638-652 Available from https://proceedings.mlr.press/v172/kanter22a.html.

Related Material