Feedback Graph Attention Convolutional Network for MR Images Enhancement by Exploring Self-Similarity Features

Xiaobin Hu, Yanyang Yan, Wenqi Ren, Hongwei Li, Amirhossein Bayat, Yu Zhao, Bjoern Menze
Proceedings of the Fourth Conference on Medical Imaging with Deep Learning, PMLR 143:327-337, 2021.

Abstract

Artifacts, blur, and noise are the common distortions degrading MRI images during the acquisition process, and deep neural networks have been demonstrated to help in improving image quality. To well exploit global structural information and self-similarity details, we propose a novel MR image enhancement network, named Feedback Graph Attention Convolutional Network (FB-GACN). As a key innovation, we consider the global structure of an image by building a graph network from image sub-regions that we consider to be node features, linking them non-locally according to their similarity. The proposed model consists of three main parts: 1) The parallel graph similarity branch and content branch, where the graph similarity branch aims at exploiting the similarity and symmetry across different image sub-regions in low-resolution feature space and provides additional priors for the content branch to enhance texture details. 2) A feedback mechanism with a recurrent structure to refine low-level representations with high-level information and generate powerful high-level texture details by handling the feedback connections. 3) A reconstruction to remove the artifacts and recover super-resolution images by using the estimated sub-region self-similarity priors obtained from the graph similarity branch. We evaluate our method on two image enhancement tasks: i) cross-protocol super resolution of diffusion MRI; ii) artifact removal of FLAIR MR images. Experimental results demonstrate that the proposed algorithm outperforms the state-of-the-art methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v143-hu21a, title = {Feedback Graph Attention Convolutional Network for {MR} Images Enhancement by Exploring Self-Similarity Features}, author = {Hu, Xiaobin and Yan, Yanyang and Ren, Wenqi and Li, Hongwei and Bayat, Amirhossein and Zhao, Yu and Menze, Bjoern}, booktitle = {Proceedings of the Fourth Conference on Medical Imaging with Deep Learning}, pages = {327--337}, year = {2021}, editor = {Heinrich, Mattias and Dou, Qi and de Bruijne, Marleen and Lellmann, Jan and Schläfer, Alexander and Ernst, Floris}, volume = {143}, series = {Proceedings of Machine Learning Research}, month = {07--09 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v143/hu21a/hu21a.pdf}, url = {https://proceedings.mlr.press/v143/hu21a.html}, abstract = {Artifacts, blur, and noise are the common distortions degrading MRI images during the acquisition process, and deep neural networks have been demonstrated to help in improving image quality. To well exploit global structural information and self-similarity details, we propose a novel MR image enhancement network, named Feedback Graph Attention Convolutional Network (FB-GACN). As a key innovation, we consider the global structure of an image by building a graph network from image sub-regions that we consider to be node features, linking them non-locally according to their similarity. The proposed model consists of three main parts: 1) The parallel graph similarity branch and content branch, where the graph similarity branch aims at exploiting the similarity and symmetry across different image sub-regions in low-resolution feature space and provides additional priors for the content branch to enhance texture details. 2) A feedback mechanism with a recurrent structure to refine low-level representations with high-level information and generate powerful high-level texture details by handling the feedback connections. 3) A reconstruction to remove the artifacts and recover super-resolution images by using the estimated sub-region self-similarity priors obtained from the graph similarity branch. We evaluate our method on two image enhancement tasks: i) cross-protocol super resolution of diffusion MRI; ii) artifact removal of FLAIR MR images. Experimental results demonstrate that the proposed algorithm outperforms the state-of-the-art methods.} }
Endnote
%0 Conference Paper %T Feedback Graph Attention Convolutional Network for MR Images Enhancement by Exploring Self-Similarity Features %A Xiaobin Hu %A Yanyang Yan %A Wenqi Ren %A Hongwei Li %A Amirhossein Bayat %A Yu Zhao %A Bjoern Menze %B Proceedings of the Fourth Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2021 %E Mattias Heinrich %E Qi Dou %E Marleen de Bruijne %E Jan Lellmann %E Alexander Schläfer %E Floris Ernst %F pmlr-v143-hu21a %I PMLR %P 327--337 %U https://proceedings.mlr.press/v143/hu21a.html %V 143 %X Artifacts, blur, and noise are the common distortions degrading MRI images during the acquisition process, and deep neural networks have been demonstrated to help in improving image quality. To well exploit global structural information and self-similarity details, we propose a novel MR image enhancement network, named Feedback Graph Attention Convolutional Network (FB-GACN). As a key innovation, we consider the global structure of an image by building a graph network from image sub-regions that we consider to be node features, linking them non-locally according to their similarity. The proposed model consists of three main parts: 1) The parallel graph similarity branch and content branch, where the graph similarity branch aims at exploiting the similarity and symmetry across different image sub-regions in low-resolution feature space and provides additional priors for the content branch to enhance texture details. 2) A feedback mechanism with a recurrent structure to refine low-level representations with high-level information and generate powerful high-level texture details by handling the feedback connections. 3) A reconstruction to remove the artifacts and recover super-resolution images by using the estimated sub-region self-similarity priors obtained from the graph similarity branch. We evaluate our method on two image enhancement tasks: i) cross-protocol super resolution of diffusion MRI; ii) artifact removal of FLAIR MR images. Experimental results demonstrate that the proposed algorithm outperforms the state-of-the-art methods.
APA
Hu, X., Yan, Y., Ren, W., Li, H., Bayat, A., Zhao, Y. & Menze, B.. (2021). Feedback Graph Attention Convolutional Network for MR Images Enhancement by Exploring Self-Similarity Features. Proceedings of the Fourth Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 143:327-337 Available from https://proceedings.mlr.press/v143/hu21a.html.

Related Material