核磁共振水分子扩散张量成像中基于广义Fibonacci数列的扩散敏感梯度磁场方向分布方案

为了准确得到人体内水分子各向异性扩散信息,在核磁共振扩散张量成像及高角分辨率扩散成像实验中,需要在众多空间均匀分布的方向上依次施加扩散敏感梯度磁场,测量水分子在不同方向上的扩散系数.目前方向分布方案的缺点有方向数目不连续、均匀性有待提高及部分方向数据的损坏会影响整个数据集等.本文以广义Fibonacci数列为基础,提出新的可以产生连续方向数目的扩散敏感梯度磁场方向分布方案,整个方案的方向均匀性较好,数据集内的部分数据仍然具有很好的空间均匀性,而且本方案中相邻两个扩散敏感梯度磁场方向接近相反,可以减小快速变化的高强度梯度磁场产生的涡流对结果的影响.

An optimal direction strategy of diffusion sensitive gradient mangnetic fields in magnetic resonance diffusion tensor imaging based on generalized Fibonacci sequence

In order to accurately investigate the directionally anisotropic diffusion information of water molecule in tissue, the diffusion sensitive gradient fields need to be applied alone many directions in order to obtain corresponding diffusion coefficients in diffusion tensor imaging (DTI) and high angular resolution diffusion imaging (HARDI) experiments. The problems facing to current diffusion sensitive gradient magnetic fields encoding schemes include the spatial uniformity of directions needs to be improved, there is no general direction design for arbitrary number of directions, flaw in any directions will cause failure or defect of the whole dataset. In this paper, we provide a generalized Fibonacci number based direction encoding scheme. This scheme can generate nearly uniform distribution for arbitrary number of directions and satisfy the spatial uniformity using partial directions from one raw data set. Besides, the diffusion sensitive gradients of neighboring directions are nearly opposite, which will reduce eddy current induced by rapid varying gradient magnetic fields.