Probing Cell Sizes Using Short Diffusion Times

一种新的磁共振成像(MRI)技术将被开发出来，能够推断具有复杂几何结构的样品中水扩散的微米级限制。在接下来的五年里，该方法将被开发用于活体受试者，并扩展到使用更精确的白质几何模型，包括取向弥散、隔室和脑脊液之间的交换。简而言之，目前的方法使用了样本几何的简化模型；在我们的例子中，我们将纤维中的轴突建模为不同直径的平行不渗透圆柱体的分布。得到了垂直于圆柱体测量的水的扩散作为频率的函数的解析表达式。将实验或蒙特卡罗模拟的数据拟合到这个分析模型中，以获得不同直径的轴突的比例。我们将改进该模型，以包括可渗透膜，以便在白质中得到更准确的结果。有了可渗透膜，水分子可以扩散的距离比轴突直径更长。如果目前的模型没有考虑到这一点，可能会扭曲结果，使轴突看起来比实际大得多。我们还将用髓鞘双层包围轴突。髓磷脂双层会增加更多的扩散屏障，其作用尚未得到充分的研究。此外，与中枢神经系统其他部分的水相比，髓鞘脂双层中的水具有非常短的松弛时间。虽然髓鞘会对信号造成更多的限制，但髓鞘双层中的水的磁化会更快地松弛，可能会达到这样的程度，即水对信号没有贡献。我们将找出模型的解析公式，并用蒙特卡罗模拟以及胼胝体和脊髓样本来检验几何和假设。我们还将优化新的方法，使其成为解决磁共振纤维束成像中接吻或交叉纤维问题的关键。磁共振波束成像使用复杂的方法，使用许多梯度方向来收集图像，以几何方式确定纤维在连接处是接吻还是交叉。测量交界处两侧纤维的轴突直径分布可以简单地确定如果两条纤维含有不同直径的轴突，纤维是接吻还是交叉。这对神经科学非常重要，因为脑束图被用来了解大脑的哪些区域是相互连接的。新方法将在交界处的梯度频率、幅度和工作方向的选择方面进行优化，目标是制作一种更准确、更快速的交界处的声道成像方法。该方法将通过白质、蔬菜和其他微米尺度的多孔样品(如水泥和肺空间)进行验证。

A new magnetic resonance imaging (MRI) technique will be developed with the capability to infer micron-scale restrictions to water diffusion in samples with complicated geometries.Over the next five years the method will be developed to work on live subjects and extended to use more accurate geometric models of white matter which include orientation dispersion, exchange between compartments and cerebrospinal fluid. In short, the current method uses a simplified model of the geometry of the sample; in our case we model axons in fibres as a distribution of parallel impermeable cylinders of varying diameters. An analytical expression is found for the diffusion of water measured perpendicular to the cylinders as a function of frequency. Data from experiments or Monte Carlo simulations are fitted to this analytical model to obtain the fraction of axons with different diameters.We will improve the model to include permeable membranes for more accurate results in white matter. With permeable membranes, water molecules can diffuse distances longer than an axon diameter. The failure of the current model to take this into account could skew the results making axons seem larger than they actually are.We will also surround the axons with myelin bilayers. Myelin lipid bilayers will add more diffusion barriers, the effect of which has not been fully studied. In addition, water within the myelin lipid bilayers has a very short relaxation time compared to water in other parts of the central nervous system. While the myelin will cause more restriction to the signal, the magnetization of the water within the myelin bilayers will relax much quicker possibly to the point that that water does not contribute to the signal. We will find analytical formulae for the model and test the geometries and assumptions with Monte Carlo simulations and corpus callosum and spinal cord samples.We will also optimize the new method so it will be a key in solving the problem with kissing or crossing fibres in MR tractography. MR tractography uses sophisticated methods collecting images using many gradient directions to determine geometrically whether fibres kiss or cross at junctions. Measuring the axon diameter distributions of fibres on either side of the junction will allow a simple determination of whether the fibres kiss or cross if the two fibres contain axons of different diameters. This is of extreme importance to neuroscience because tractography is being used to understand which regions of the brain are connected to each other. The new method will be optimized in terms of the selection of gradient frequencies, amplitudes, and directions to work at junctions with the goal of making a more accurate and faster method for tractography at junctions.The method will be verified using white matter, vegetables, and other micron-scale porous samples such as cement and lung airspaces.