Rethinking Dual-Domain Undersampled MRI reconstruction: domain-specific design from the perspective of the receptive field
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913234651447296 |
|---|---|
| author | Gao, Ziqi Zhou, S. Kevin |
| author_facet | Gao, Ziqi Zhou, S. Kevin |
| contents | Undersampled MRI reconstruction is crucial for accelerating clinical scanning. Dual-domain reconstruction network is performant among SoTA deep learning methods. In this paper, we rethink dual-domain model design from the perspective of the receptive field, which is needed for image recovery and K-space interpolation problems. Further, we introduce domain-specific modules for dual-domain reconstruction, namely k-space global initialization and image-domain parallel local detail enhancement. We evaluate our modules by translating a SoTA method DuDoRNet under different conventions of MRI reconstruction including image-domain, dual-domain, and reference-guided reconstruction on the public IXI dataset. Our model DuDoRNet+ achieves significant improvements over competing deep learning methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2303_10611 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Rethinking Dual-Domain Undersampled MRI reconstruction: domain-specific design from the perspective of the receptive field Gao, Ziqi Zhou, S. Kevin Image and Video Processing Computer Vision and Pattern Recognition Undersampled MRI reconstruction is crucial for accelerating clinical scanning. Dual-domain reconstruction network is performant among SoTA deep learning methods. In this paper, we rethink dual-domain model design from the perspective of the receptive field, which is needed for image recovery and K-space interpolation problems. Further, we introduce domain-specific modules for dual-domain reconstruction, namely k-space global initialization and image-domain parallel local detail enhancement. We evaluate our modules by translating a SoTA method DuDoRNet under different conventions of MRI reconstruction including image-domain, dual-domain, and reference-guided reconstruction on the public IXI dataset. Our model DuDoRNet+ achieves significant improvements over competing deep learning methods. |
| title | Rethinking Dual-Domain Undersampled MRI reconstruction: domain-specific design from the perspective of the receptive field |
| topic | Image and Video Processing Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2303.10611 |