Deep learning based parallel Imaging (PI) has made great progresses in recent years to accelerate magnetic resonance imaging (MRI). Nevertheless, the performanc-es and robustness of existing methods can still be im-proved. In this work, we propose to explore the k-space domain learning via robust generative modeling for flexible PI reconstruction, coined weight-k-space genera-tive model (WKGM). Specifically, WKGM is a general-ized k-space domain model, where the k-space weighting technology and high-dimensional space augmentation design are efficiently incorporated for score-based gen-erative model training, resulting in good and robust re-constructions. In addition, WKGM is flexible and thus can be synergistically combined with various traditional k-space PI models, generating learning-based priors to produce high-fidelity reconstructions. Experimental re-sults on datasets with varying sampling patterns and ac-celeration factors demonstrate that WKGM can attain state-of-the-art reconstruction results with the well-learned k-space generative prior.

翻译：近年来,在加速磁共振成像(MRI)方面,深层学习平行成像(PI)取得了巨大进展。然而,现有方法的性能和稳健性仍然可以证明。在这项工作中,我们提议探索 k-空间域学习,为此,为灵活的PI重建建立强大的基因化模型,创建重量-k-空间基因动力模型(WKGM),创建基于学习的先期模型,以产生高纤维性重建。具体地说,WKGM是一个通用的k-空间域模型,其中K-空间加权技术和高维空间空间增强设计被有效地纳入基于分的基因模型培训,从而产生良好和强健的再造。此外,WKGM具有灵活性,因此可以与各种传统的K-空间PI模型协同,生成基于学习的先期模型,以产生高纤维性重建。在具有不同取样模式和加速因素的数据集上进行实验性再解,表明WKKGM能够取得先进的重建结果。