Laplacian pyramid-based complex neural network learning for fast MR imaging

Haoyun Liang, Yu Gong, Hoel Kervadec, Cheng Li, Jing Yuan, Xin Liu, Hairong Zheng, Shanshan Wang
Proceedings of the Third Conference on Medical Imaging with Deep Learning, PMLR 121:454-464, 2020.

Abstract

A Laplacian pyramid-based complex neural network, CLP-Net, is proposed to reconstruct high-quality magnetic resonance images from undersampled k-space data. Specifically, three major contributions have been made: 1) A new framework has been proposed to explore the encouraging multi-scale properties of Laplacian pyramid decomposition; 2) A cascaded multi-scale network architecture with complex convolutions has been designed under the proposed framework; 3) Experimental validations on an open source dataset fastMRI demonstrate the encouraging properties of the proposed method in preserving image edges and fine textures.

Cite this Paper

BibTeX
@InProceedings{pmlr-v121-liang20a, title = {Laplacian pyramid-based complex neural network learning for fast MR imaging}, author = {Liang, Haoyun and Gong, Yu and Kervadec, Hoel and Li, Cheng and Yuan, Jing and Liu, Xin and Zheng, Hairong and Wang, Shanshan}, booktitle = {Proceedings of the Third Conference on Medical Imaging with Deep Learning}, pages = {454--464}, year = {2020}, editor = {Arbel, Tal and Ben Ayed, Ismail and de Bruijne, Marleen and Descoteaux, Maxime and Lombaert, Herve and Pal, Christopher}, volume = {121}, series = {Proceedings of Machine Learning Research}, month = {06--08 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v121/liang20a/liang20a.pdf}, url = {https://proceedings.mlr.press/v121/liang20a.html}, abstract = {A Laplacian pyramid-based complex neural network, CLP-Net, is proposed to reconstruct high-quality magnetic resonance images from undersampled k-space data. Specifically, three major contributions have been made: 1) A new framework has been proposed to explore the encouraging multi-scale properties of Laplacian pyramid decomposition; 2) A cascaded multi-scale network architecture with complex convolutions has been designed under the proposed framework; 3) Experimental validations on an open source dataset fastMRI demonstrate the encouraging properties of the proposed method in preserving image edges and fine textures.} }
Endnote
%0 Conference Paper %T Laplacian pyramid-based complex neural network learning for fast MR imaging %A Haoyun Liang %A Yu Gong %A Hoel Kervadec %A Cheng Li %A Jing Yuan %A Xin Liu %A Hairong Zheng %A Shanshan Wang %B Proceedings of the Third Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2020 %E Tal Arbel %E Ismail Ben Ayed %E Marleen de Bruijne %E Maxime Descoteaux %E Herve Lombaert %E Christopher Pal %F pmlr-v121-liang20a %I PMLR %P 454--464 %U https://proceedings.mlr.press/v121/liang20a.html %V 121 %X A Laplacian pyramid-based complex neural network, CLP-Net, is proposed to reconstruct high-quality magnetic resonance images from undersampled k-space data. Specifically, three major contributions have been made: 1) A new framework has been proposed to explore the encouraging multi-scale properties of Laplacian pyramid decomposition; 2) A cascaded multi-scale network architecture with complex convolutions has been designed under the proposed framework; 3) Experimental validations on an open source dataset fastMRI demonstrate the encouraging properties of the proposed method in preserving image edges and fine textures.
APA
Liang, H., Gong, Y., Kervadec, H., Li, C., Yuan, J., Liu, X., Zheng, H. & Wang, S.. (2020). Laplacian pyramid-based complex neural network learning for fast MR imaging. Proceedings of the Third Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 121:454-464 Available from https://proceedings.mlr.press/v121/liang20a.html.

Related Material