RF Encoding for Gradient-Free MRI

描述（由申请人提供）：本项目的目标是开发新的射频（RF）梯度编码方法来对MRI信号进行空间编码。这些方法将使无声、低成本的MRI系统成为可能，从而大幅降低成像成本，提高患者的依从性和舒适度。在传统的MRI中，接收到的信号根据其时间频率定位到其空间原点，该频率由平行于扫描仪主场（B0）并在空间上线性变化的磁场控制。这些B0梯度场存在许多问题：它们声音很大，会刺激周围神经，影响患者的舒适度；由于线圈的高电感，它们具有相对较长的开关时间；它们需要庞大的冷却系统和定制的放大器；而且它们很贵，约占临床扫描仪成本的20-25%。这些问题的一个潜在解决方案是用RF梯度代替B0梯度，这是静音和低成本的。不幸的是，尽管其潜在的射频梯度编码尚未成为临床或商业上的成功。这在很大程度上是由于现有的RF梯度编码方法没有提供B0梯度所享有的对比度开发和空间编码之间的正交性，或者将现有的基于B0梯度的MRI采集技术转换为使用RF编码的直接途径。本项目中提出的方法将是第一个满足这些要求的方法，因此将代表第一个真正可行的基于射频梯度的成像方法。这个项目的核心创新是使用Bloch-Siegert变换对MRI信号进行空间编码。与B0梯度一样，这种编码机制基于直接在横向平面上应用相移来磁化，因此不会调制横向磁化的大小，从而使图像对比度不受空间编码的影响。该项目的第一个目标是开发新的射频梯度线圈和其他射频硬件，以实现0.5特斯拉的2D和3D笛卡尔成像，包括同时射频传输和接收的硬件策略，以实现布洛赫-西格特移位的频率编码。第二个目标是开发新的基于Bloch-Siegert变换的射频编码脉冲序列，利用最近在Bloch-Siegert相位编码的射频脉冲设计和基于射频梯度的切片选择激励的射频脉冲设计方面的创新。第三个目标是开发鲁棒算法，从rf编码数据重建图像。这些目标的成功完成将广泛证明所提出的射频编码方法用于磁共振成像的可行性，为翻译到人类铺平道路。

DESCRIPTION (provided by applicant): The goal of this project is to develop new radiofrequency (RF) gradient encoding methods to spatially encode signals in MRI. The methods would enable silent, low-cost MRI systems, leading to a substantial reduction in the cost of imaging and improved patient compliance and comfort. In conventional MRI, a received signal is localized to its spatial location of origin based on its temporal frequency, which is controlled using magnetic fields that are parallel to the main (B0) field of the scanner and vary linearly across space. There are many problems with these B0 gradient fields: they are loud and induce peripheral nerve stimulation, compromising patient comfort; they have relatively long switching times due to the high inductance of the coils; they require bulky cooling systems and customized amplifiers; and they are expensive, representing around 20-25% of the cost of a clinical scanner. A potential solution to these problems is to replace B0 gradients with RF gradients, which are silent and low-cost. Unfortunately, in spite of its potential RF gradient encoding has not yet become a clinical or commercial success. This is largely due to the fact that no existing RF gradient encoding method offers the orthogonality between contrast development and spatial encoding that is en- joyed by B0 gradients, or a straightforward path to convert existing B0 gradient-based MRI acquisition techniques to use RF encoding. The methods proposed in this project will be the first to meet these requirements, and will thus represent the first truly viable RF gradient-based imaging methods. The central innovation of this project is to use the Bloch-Siegert shift to spatially encode the MRI signal. As with B0 gradients, this encoding mechanism is based on the application of phase shifts to magnetization directly in the transverse plane, and therefore does not modulate the magnitude of the transverse magnetization, leaving image contrast unaffected by spatial encoding. The first Aim of the project is to develop new RF gradient coils and other RF hardware to enable 2D and 3D Cartesian imaging at 0.5 Tesla, including hardware strategies for simultaneous RF transmission and reception to enable frequency encoding by Bloch-Siegert shift. The second Aim is to develop new RF-encoded pulse sequences based on the Bloch-Siegert shift, leveraging recent innovations in RF pulse design for Bloch-Siegert phase encoding, and in RF pulse design for RF gradient-based slice-selective excitation. The third Aim is to develop robust algorithms to reconstruct images from RF-encoded data. Successful completion of these Aims would broadly prove the feasibility of the proposed RF encoding methods for MR imaging, paving the way for translation to humans.