Center for Biomedical Imaging Research, Department of Biomedical Engineering, School of Medicine, Tsinghua University, Beijing, China; Department of Radiology, Stanford University, Stanford, CA, United States.
Center for Biomedical Imaging Research, Department of Biomedical Engineering, School of Medicine, Tsinghua University, Beijing, China.
Neuroimage. 2021 Aug 15;237:118099. doi: 10.1016/j.neuroimage.2021.118099. Epub 2021 May 1.
High-resolution diffusion MRI (dMRI) is a crucial tool in neuroscience studies to detect fine fiber structure, depict complex fiber architecture and analyze cortical anisotropy. However, high-resolution dMRI is limited by its intrinsically low SNR due to diffusion attenuation. A series of techniques have been proposed to improve the SNR performance, but most of them are at the cost of long scan time, which in turn sacrifice the SNR efficiency, especially for large FOV imaging, such as whole-brain imaging. Recently, a combination of 3D multi-slab acquisition and simultaneous multi-slice (SMS) excitation, namely simultaneous multi-slab (SMSlab), has been demonstrated to have potential for high-resolution diffusion imaging with high SNR and SNR efficiency. In our previous work, we have proposed a 3D Fourier encoding and reconstruction framework for SMSlab acquisition. In this study, we extend this 3D k-space framework to diffusion imaging, by developing a novel navigator acquisition strategy and exploring a k-space-based phase correction method. In vivo brain data are acquired using the proposed SMSlab method and compared with a series of different acquisitions, including the traditional 3D multi-slab, 2D SMS and 2D single-shot EPI (ss-EPI) acquisitions. The results demonstrate that SMSlab has a better SNR performance compared with 3D multi-slab and 2D SMS. The detection capacity of fine fiber structures is improved using SMSlab, compared with the low-resolution diffusion imaging using conventional 2D ss-EPI.
高分辨率弥散磁共振成像（dMRI）是神经科学研究中检测精细纤维结构、描绘复杂纤维结构和分析皮质各向异性的重要工具。然而，由于弥散衰减，高分辨率 dMRI 的固有信噪比受到限制。已经提出了一系列提高 SNR 性能的技术，但它们大多数都以长扫描时间为代价，这反过来又牺牲了 SNR 效率，特别是对于大视野成像，如全脑成像。最近，3D 多切片采集和同时多切片（SMS）激发的组合，即同时多切片（SMSlab），已被证明具有高 SNR 和 SNR 效率的高分辨率弥散成像的潜力。在我们之前的工作中，我们已经提出了一种用于 SMSlab 采集的 3D 傅里叶编码和重建框架。在这项研究中，我们通过开发一种新的导航采集策略和探索基于 k 空间的相位校正方法，将这个 3D k 空间框架扩展到扩散成像中。使用所提出的 SMSlab 方法采集了活体大脑数据，并与一系列不同的采集方法进行了比较，包括传统的 3D 多切片、2D SMS 和 2D 单次激发 EPI（ss-EPI）采集。结果表明，与 3D 多切片和 2D SMS 相比，SMSlab 具有更好的 SNR 性能。与传统的 2D ss-EPI 低分辨率扩散成像相比，SMSlab 提高了精细纤维结构的检测能力。
