TY - GEN
T1 - SRE-CNN
T2 - 27th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2024
AU - Zhu, Yuliang
AU - Cheng, Jing
AU - Cui, Zhuo Xu
AU - Ren, Jianfeng
AU - Wang, Chengbo
AU - Liang, Dong
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.
PY - 2024
Y1 - 2024
N2 - Dynamic MR images possess various transformation symmetries, including the rotation symmetry of local features within the image and along the temporal dimension. Utilizing these symmetries as prior knowledge can facilitate dynamic MR imaging with high spatiotemporal resolution. Equivariant CNN is an effective tool to leverage the symmetry priors. However, current equivariant CNN methods fail to fully exploit these symmetry priors in dynamic MR imaging. In this work, we propose a novel framework of Spatiotemporal Rotation-Equivariant CNN (SRE-CNN), spanning from the underlying high-precision filter design to the construction of the temporal-equivariant convolutional module and imaging model, to fully harness the rotation symmetries inherent in dynamic MR images. The temporal-equivariant convolutional module enables exploitation the rotation symmetries in both spatial and temporal dimensions, while the high-precision convolutional filter, based on parametrization strategy, enhances the utilization of rotation symmetry of local features to improve the reconstruction of detailed anatomical structures. Experiments conducted on highly undersampled dynamic cardiac cine data (up to 20X) have demonstrated the superior performance of our proposed approach, both quantitatively and qualitatively.
AB - Dynamic MR images possess various transformation symmetries, including the rotation symmetry of local features within the image and along the temporal dimension. Utilizing these symmetries as prior knowledge can facilitate dynamic MR imaging with high spatiotemporal resolution. Equivariant CNN is an effective tool to leverage the symmetry priors. However, current equivariant CNN methods fail to fully exploit these symmetry priors in dynamic MR imaging. In this work, we propose a novel framework of Spatiotemporal Rotation-Equivariant CNN (SRE-CNN), spanning from the underlying high-precision filter design to the construction of the temporal-equivariant convolutional module and imaging model, to fully harness the rotation symmetries inherent in dynamic MR images. The temporal-equivariant convolutional module enables exploitation the rotation symmetries in both spatial and temporal dimensions, while the high-precision convolutional filter, based on parametrization strategy, enhances the utilization of rotation symmetry of local features to improve the reconstruction of detailed anatomical structures. Experiments conducted on highly undersampled dynamic cardiac cine data (up to 20X) have demonstrated the superior performance of our proposed approach, both quantitatively and qualitatively.
KW - Cardiac MR image reconstruction
KW - Filter parametrization
KW - Rotation equivariant convolution
KW - Spatiotemporal equivariance
UR - http://www.scopus.com/inward/record.url?scp=85212501719&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-72104-5_65
DO - 10.1007/978-3-031-72104-5_65
M3 - Conference contribution
AN - SCOPUS:85212501719
SN - 9783031721038
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 679
EP - 689
BT - Medical Image Computing and Computer Assisted Intervention – MICCAI 2024 - 27th International Conference, Proceedings
A2 - Linguraru, Marius George
A2 - Dou, Qi
A2 - Feragen, Aasa
A2 - Giannarou, Stamatia
A2 - Glocker, Ben
A2 - Lekadir, Karim
A2 - Schnabel, Julia A.
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 6 October 2024 through 10 October 2024
ER -