BibTex format
@article{Wang:2025:10.1109/TMI.2025.3641610,
author = {Wang, F and Wang, Z and Li, Y and Lyu, J and Qin, C and Wang, S and Guo, K and Sun, M and Huang, M and Zhang, H and Tanzer, M and Li, Q and Chen, X and Huang, J and Wu, Y and Zhang, H and Hamedani, KA and Lyu, Y and Sun, L and Li, Q and He, T and Lan, L and Yao, Q and Xu, Z and Xin, B and Metaxas, DN and Razizadeh, N and Nabavi, S and Yiasemis, G and Teuwen, J and Zhang, Z and Wang, S and Zhang, C and Ennis, DB and Xue, Z and Hu, C and Xu, R and Oksuz, I and Lyu, D and Huang, Y and Guo, X and Hao, R and Patel, JH and Cai, G and Chen, B and Zhang, Y and Hua, S and Chen, Z and Dou, Q and Zhuang, X and Tao, Q and Bai, W and Qin, J and Wang, H and Prieto, C and Markl, M and Young, A and Li, H and Hu, X and Wu, L and Qu, X and Yang, G and Wang, C},
doi = {10.1109/TMI.2025.3641610},
journal = {IEEE Trans Med Imaging},
title = {Towards Modality- and Sampling-Universal Learning Strategies for Accelerating Cardiovascular Imaging: Summary of the CMRxRecon2024 Challenge.},
url = {http://dx.doi.org/10.1109/TMI.2025.3641610},
volume = {PP},
year = {2025}
}
RIS format (EndNote, RefMan)
TY - JOUR
AB - Cardiovascular health is vital to human well-being, and cardiac magnetic resonance (CMR) imaging is considered the clinical reference standard for diagnosing cardiovascular disease. However, its adoption is hindered by long scan times, complex contrasts, and inconsistent quality. While deep learning methods perform well on specific CMR imaging sequences, they often fail to generalize across modalities and sampling schemes. The lack of benchmarks for high-quality, fast CMR image reconstruction further limits technology comparison and adoption. The CMRxRecon2024 challenge, attracting over 200 teams from 18 countries, addressed these issues with two tasks: generalization to unseen modalities and robustness to diverse undersampling patterns. We introduced the largest public multi-modality CMR raw dataset, an open benchmarking platform, and shared code. Analysis of the best-performing solutions revealed that prompt-based adaptation and enhanced physics-driven consistency enabled strong cross-scenario performance. These findings establish principles for generalizable reconstruction models and advance clinically translatable AI in cardiovascular imaging.
AU - Wang,F
AU - Wang,Z
AU - Li,Y
AU - Lyu,J
AU - Qin,C
AU - Wang,S
AU - Guo,K
AU - Sun,M
AU - Huang,M
AU - Zhang,H
AU - Tanzer,M
AU - Li,Q
AU - Chen,X
AU - Huang,J
AU - Wu,Y
AU - Zhang,H
AU - Hamedani,KA
AU - Lyu,Y
AU - Sun,L
AU - Li,Q
AU - He,T
AU - Lan,L
AU - Yao,Q
AU - Xu,Z
AU - Xin,B
AU - Metaxas,DN
AU - Razizadeh,N
AU - Nabavi,S
AU - Yiasemis,G
AU - Teuwen,J
AU - Zhang,Z
AU - Wang,S
AU - Zhang,C
AU - Ennis,DB
AU - Xue,Z
AU - Hu,C
AU - Xu,R
AU - Oksuz,I
AU - Lyu,D
AU - Huang,Y
AU - Guo,X
AU - Hao,R
AU - Patel,JH
AU - Cai,G
AU - Chen,B
AU - Zhang,Y
AU - Hua,S
AU - Chen,Z
AU - Dou,Q
AU - Zhuang,X
AU - Tao,Q
AU - Bai,W
AU - Qin,J
AU - Wang,H
AU - Prieto,C
AU - Markl,M
AU - Young,A
AU - Li,H
AU - Hu,X
AU - Wu,L
AU - Qu,X
AU - Yang,G
AU - Wang,C
DO - 10.1109/TMI.2025.3641610
PY - 2025///
TI - Towards Modality- and Sampling-Universal Learning Strategies for Accelerating Cardiovascular Imaging: Summary of the CMRxRecon2024 Challenge.
T2 - IEEE Trans Med Imaging
UR - http://dx.doi.org/10.1109/TMI.2025.3641610
UR - https://www.ncbi.nlm.nih.gov/pubmed/41359736
VL - PP
ER -