发表机构
School of Integrated Circuits and Electronics, Beijing Institute of Technology; School of Biomedical Engineering, Tsinghua Medicine, Tsinghua University; Department of Electronic Engineering, The Chinese University of Hong Kong（北京理工大学集成电路与电子学院; 清华大学医学院生物医学工程学院; 香港中文大学电子工程系）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对跨场MRI转换的域偏移问题，提出基于场条件内容-风格预训练的3D非配对转换框架，适配三项挑战赛任务，可保留三维解剖结构。
AI 中文摘要
磁场强度是磁共振成像（MRI）中域偏移的主要来源，会影响信噪比、组织对比度、空间细节以及解剖边界的可见性。MRIxFields 2026挑战赛研究了这一问题，涉及0.1T、1.5T、3T、5T和7T不同场强采集的跨场MRI转换，其三项任务（任意-to-7T、0.1T-to-High、任意-to-任意合成）要求生成目标场图像特征，同时保留受试者特定的解剖结构。该问题极具挑战性，因为训练时很少能获得同一受试者在多个场强下的配对采集数据。我们提出一种基于场条件内容-风格预训练的3D非配对跨场MRI转换框架，该框架首先通过将解剖内容与场依赖的对比度特征解耦，学习所有可用场强下可控的场间转换，随后将预训练的骨干网络适配到任务特定的目标域。我们的模型包含3D内容编码器、3D风格编码器、场条件风格生成器、AdaIN调制解码器以及多场判别器；对抗学习鼓励生成逼真的目标场外观，而循环一致性、身份、内容、风格和多样性约束则促进解剖保真度与可控转换。我们在涵盖5种场强和3种MRI模态的MRIxFields数据上评估该方法，对配对测试数据的实验表明，该框架可适配三项挑战赛设置，同时在合成 volumes 中保留三维解剖结构。实现代码公开于此https URL。
英文摘要
Magnetic field strength is a major source of domain shift in magnetic resonance imaging (MRI), affecting signal-to-noise ratio, tissue contrast, spatial detail, and the visibility of anatomical boundaries. The MRIxFields 2026 challenge investigates this problem through cross-field MRI translation across acquisitions at 0.1T, 1.5T, 3T, 5T, and 7T. Its three tasks, Any-to-7T, 0.1T-to-High, and Any-to-Any synthesis, require the generation of target-field image characteristics while preserving subject-specific anatomy. This problem is particularly challenging because paired acquisitions of the same subject across multiple field strengths are rarely available for training. We propose a 3D unpaired cross-field MRI translation framework based on field-conditioned content-style pretraining. The proposed framework first learns controllable field-to-field translation across all available field strengths by disentangling anatomical content from field-dependent contrast characteristics. The pretrained backbone is then adapted to task-specific target domains. Our model comprises a 3D content encoder, a 3D style encoder, a field-conditioned style generator, an AdaIN-modulated decoder, and a multi-field discriminator. Adversarial learning encourages realistic target-field appearance, while cycle-consistency, identity, content, style, and diversity constraints promote anatomical fidelity and controllable translation. We evaluate the proposed method on MRIxFields data spanning five field strengths and three MRI modalities. Experiments on paired test data demonstrate that the framework can adapt to the three challenge settings while preserving three-dimensional anatomical structure in the synthesized volumes. The implementation code is publicly available at https://github.com/Idea89560041/3D-MRI-Field-Translation.
CommentsMICCAI 2026 Workshop on MRIxFields
