SuperMask: Generating High-resolution object masks from multi-view, unaligned low-resolution MRIs

Hanxue Gu, Hongyu He, Roy Colglazier, Jordan Axelrod, Robert French, Maciej A Mazurowski
Medical Imaging with Deep Learning, PMLR 227:119-133, 2024.

Abstract

Three-dimensional segmentation in magnetic resonance images (MRI), which reflects the true shape of the objects, is challenging since high-resolution isotropic MRIs are rare and typical MRIs are anisotropic, with the out-of-plane dimension having a much lower resolution. A potential remedy to this issue lies in the fact that often multiple sequences are acquired on different planes. However, in practice, these sequences are not orthogonal to each other, limiting the applicability of many previous solutions to reconstruct higher-resolution images from multiple lower-resolution ones. We propose a novel deep learning-based solution to generating high-resolution masks from multiple low-resolution images. Our method combines segmentation and unsupervised registration networks by introducing two new regularizations to make registration and segmentation reinforce each other. Finally, we introduce a multi-view fusion method to generate high-resolution target object masks. The experimental results on two datasets show the superiority of our methods. Importantly, the advantage of not using high-resolution images in the training process makes our method applicable to a wide variety of MRI segmentation tasks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v227-gu24b, title = {SuperMask: Generating High-resolution object masks from multi-view, unaligned low-resolution MRIs}, author = {Gu, Hanxue and He, Hongyu and Colglazier, Roy and Axelrod, Jordan and French, Robert and Mazurowski, Maciej A}, booktitle = {Medical Imaging with Deep Learning}, pages = {119--133}, year = {2024}, editor = {Oguz, Ipek and Noble, Jack and Li, Xiaoxiao and Styner, Martin and Baumgartner, Christian and Rusu, Mirabela and Heinmann, Tobias and Kontos, Despina and Landman, Bennett and Dawant, Benoit}, volume = {227}, series = {Proceedings of Machine Learning Research}, month = {10--12 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v227/gu24b/gu24b.pdf}, url = {https://proceedings.mlr.press/v227/gu24b.html}, abstract = {Three-dimensional segmentation in magnetic resonance images (MRI), which reflects the true shape of the objects, is challenging since high-resolution isotropic MRIs are rare and typical MRIs are anisotropic, with the out-of-plane dimension having a much lower resolution. A potential remedy to this issue lies in the fact that often multiple sequences are acquired on different planes. However, in practice, these sequences are not orthogonal to each other, limiting the applicability of many previous solutions to reconstruct higher-resolution images from multiple lower-resolution ones. We propose a novel deep learning-based solution to generating high-resolution masks from multiple low-resolution images. Our method combines segmentation and unsupervised registration networks by introducing two new regularizations to make registration and segmentation reinforce each other. Finally, we introduce a multi-view fusion method to generate high-resolution target object masks. The experimental results on two datasets show the superiority of our methods. Importantly, the advantage of not using high-resolution images in the training process makes our method applicable to a wide variety of MRI segmentation tasks.} }
Endnote
%0 Conference Paper %T SuperMask: Generating High-resolution object masks from multi-view, unaligned low-resolution MRIs %A Hanxue Gu %A Hongyu He %A Roy Colglazier %A Jordan Axelrod %A Robert French %A Maciej A Mazurowski %B Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2024 %E Ipek Oguz %E Jack Noble %E Xiaoxiao Li %E Martin Styner %E Christian Baumgartner %E Mirabela Rusu %E Tobias Heinmann %E Despina Kontos %E Bennett Landman %E Benoit Dawant %F pmlr-v227-gu24b %I PMLR %P 119--133 %U https://proceedings.mlr.press/v227/gu24b.html %V 227 %X Three-dimensional segmentation in magnetic resonance images (MRI), which reflects the true shape of the objects, is challenging since high-resolution isotropic MRIs are rare and typical MRIs are anisotropic, with the out-of-plane dimension having a much lower resolution. A potential remedy to this issue lies in the fact that often multiple sequences are acquired on different planes. However, in practice, these sequences are not orthogonal to each other, limiting the applicability of many previous solutions to reconstruct higher-resolution images from multiple lower-resolution ones. We propose a novel deep learning-based solution to generating high-resolution masks from multiple low-resolution images. Our method combines segmentation and unsupervised registration networks by introducing two new regularizations to make registration and segmentation reinforce each other. Finally, we introduce a multi-view fusion method to generate high-resolution target object masks. The experimental results on two datasets show the superiority of our methods. Importantly, the advantage of not using high-resolution images in the training process makes our method applicable to a wide variety of MRI segmentation tasks.
APA
Gu, H., He, H., Colglazier, R., Axelrod, J., French, R. & Mazurowski, M.A.. (2024). SuperMask: Generating High-resolution object masks from multi-view, unaligned low-resolution MRIs. Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 227:119-133 Available from https://proceedings.mlr.press/v227/gu24b.html.

Related Material