用分割引导对抗学习，提升64毫特斯拉超低场核磁图像质量。
Enhancing Ultra-low-field MRI with Segmentation-guided Adversarial Learning
- 基于分割图先验，用CycleGAN和Transformer模型分别增强图像。
- 在64毫特斯拉数据上合成出接近3特斯拉场强的清晰图像。
- 适合医疗影像增强、便携式核磁研发人员参考。
超低场（ULF）核磁共振成像具有便携性和低成本优势，但图像质量较差。为解决此问题，我们提交了2025年超低场增强挑战赛（ULF-EnC）的方案，目标是从64毫特斯拉扫描数据中合成接近高场强的MR图像。我们的流程结合解剖结构先验与模型集成：首先使用仅依赖比赛提供数据训练的Swin UNETR生成组织分割图作为先验；随后，两个独立的增强网络——CycleGAN与基于Transformer的残差增强模型（T-REX）——分别以该先验条件化，训练生成3特斯拉类图像；最终通过加权平均融合两者输出。实验表明，该方法生成的图像在定量与定性评估上均达到接近高场扫描的效果。
原文摘要 · Abstract (English)
Ultra-low-field (ULF) MRI offers portable and low-cost imaging but suffers from poor image quality. To address this, we present our submission to the 2025 ULF Enhancement Challenge (ULF-EnC), where the goal is to synthesise high-field-like MRIs from 64 mT scans. Our pipeline enhances ULF MRI through a combination of anatomical conditioning and model ensembling. We first generate tissue segmentation priors using a Swin UNETR trained solely on challenge-provided data. These priors condition two independent enhancement networks - a CycleGAN and a transformer-based residual enhancement model (T-REX) - each trained to synthesise 3 T-like MRIs. Outputs from both models are combined using a weighted average. Our approach produces enhanced MRIs that were comparable to high-field scans both quantitatively and qualitatively.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。