Towards Modality- and Sampling-Universal Learning Strategies for Accelerating Cardiovascular Imaging: Summary of the CMRxRecon2024 Challenge
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908692160446464 |
|---|---|
| author | Wang, Fanwen Wang, Zi Li, Yan Lyu, Jun Qin, Chen Wang, Shuo Guo, Kunyuan Sun, Mengting Huang, Mingkai Zhang, Haoyu Tänzer, Michael Li, Qirong Chen, Xinran Huang, Jiahao Wu, Yinzhe Zhang, Haosen Hamedani, Kian Anvari Lyu, Yuntong Sun, Longyu Li, Qing He, Tianxing Lan, Lizhen Yao, Qiong Xu, Ziqiang Xin, Bingyu Metaxas, Dimitris N. Razizadeh, Narges Nabavi, Shahabedin Yiasemis, George Teuwen, Jonas Zhang, Zhenxi Wang, Sha Zhang, Chi Ennis, Daniel B. Xue, Zhihao Hu, Chenxi Xu, Ruru Oksuz, Ilkay Lyu, Donghang Huang, Yanxin Guo, Xinrui Hao, Ruqian Patel, Jaykumar H. Cai, Guanke Chen, Binghua Zhang, Yajing Hua, Sha Chen, Zhensen Dou, Qi Zhuang, Xiahai Tao, Qian Bai, Wenjia Qin, Jing Wang, He Prieto, Claudia Markl, Michael Young, Alistair Li, Hao Hu, Xihong Wu, Lianming Qu, Xiaobo Yang, Guang Wang, Chengyan |
| author_facet | Wang, Fanwen Wang, Zi Li, Yan Lyu, Jun Qin, Chen Wang, Shuo Guo, Kunyuan Sun, Mengting Huang, Mingkai Zhang, Haoyu Tänzer, Michael Li, Qirong Chen, Xinran Huang, Jiahao Wu, Yinzhe Zhang, Haosen Hamedani, Kian Anvari Lyu, Yuntong Sun, Longyu Li, Qing He, Tianxing Lan, Lizhen Yao, Qiong Xu, Ziqiang Xin, Bingyu Metaxas, Dimitris N. Razizadeh, Narges Nabavi, Shahabedin Yiasemis, George Teuwen, Jonas Zhang, Zhenxi Wang, Sha Zhang, Chi Ennis, Daniel B. Xue, Zhihao Hu, Chenxi Xu, Ruru Oksuz, Ilkay Lyu, Donghang Huang, Yanxin Guo, Xinrui Hao, Ruqian Patel, Jaykumar H. Cai, Guanke Chen, Binghua Zhang, Yajing Hua, Sha Chen, Zhensen Dou, Qi Zhuang, Xiahai Tao, Qian Bai, Wenjia Qin, Jing Wang, He Prieto, Claudia Markl, Michael Young, Alistair Li, Hao Hu, Xihong Wu, Lianming Qu, Xiaobo Yang, Guang Wang, Chengyan |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_03971 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Towards Modality- and Sampling-Universal Learning Strategies for Accelerating Cardiovascular Imaging: Summary of the CMRxRecon2024 Challenge Wang, Fanwen Wang, Zi Li, Yan Lyu, Jun Qin, Chen Wang, Shuo Guo, Kunyuan Sun, Mengting Huang, Mingkai Zhang, Haoyu Tänzer, Michael Li, Qirong Chen, Xinran Huang, Jiahao Wu, Yinzhe Zhang, Haosen Hamedani, Kian Anvari Lyu, Yuntong Sun, Longyu Li, Qing He, Tianxing Lan, Lizhen Yao, Qiong Xu, Ziqiang Xin, Bingyu Metaxas, Dimitris N. Razizadeh, Narges Nabavi, Shahabedin Yiasemis, George Teuwen, Jonas Zhang, Zhenxi Wang, Sha Zhang, Chi Ennis, Daniel B. Xue, Zhihao Hu, Chenxi Xu, Ruru Oksuz, Ilkay Lyu, Donghang Huang, Yanxin Guo, Xinrui Hao, Ruqian Patel, Jaykumar H. Cai, Guanke Chen, Binghua Zhang, Yajing Hua, Sha Chen, Zhensen Dou, Qi Zhuang, Xiahai Tao, Qian Bai, Wenjia Qin, Jing Wang, He Prieto, Claudia Markl, Michael Young, Alistair Li, Hao Hu, Xihong Wu, Lianming Qu, Xiaobo Yang, Guang Wang, Chengyan Image and Video Processing 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. |
| title | Towards Modality- and Sampling-Universal Learning Strategies for Accelerating Cardiovascular Imaging: Summary of the CMRxRecon2024 Challenge |
| topic | Image and Video Processing |
| url | https://arxiv.org/abs/2503.03971 |