Department of Electrical and Computer Engineering, University of Rochester, Rochester, New York.
Department of Imaging Sciences, University of Rochester, Rochester, New York.
NMR Biomed. 2022 Feb;35(2):e4628. doi: 10.1002/nbm.4628. Epub 2021 Oct 12.
Neurite orientation dispersion and density imaging (NODDI) enables the assessment of intracellular, extracellular, and free water signals from multi-shell diffusion MRI data. It is an insightful approach to characterize brain tissue microstructure. Single-shell reconstruction for NODDI parameters has been discouraged in previous studies caused by failure when fitting, especially for the neurite density index (NDI). Here, we investigated the possibility of creating robust NODDI parameter maps with single-shell data, using the isotropic volume fraction (f ) as a prior. Prior estimation was made independent of the NODDI model constraint using a dictionary learning approach. First, we used a stochastic sparse dictionary-based network (DictNet), which is trained with data obtained from in vivo and simulated diffusion MRI data, to predict f . In single-shell cases, the mean diffusivity and raw T signal with no diffusion weighting (S ) was incorporated in the dictionary for the f estimation. Then, the NODDI framework was used with the known f to estimate the NDI and orientation dispersion index (ODI). The f estimated using our model was compared with other f estimators in the simulation. Further, using both synthetic data simulation and human data collected on a 3 T scanner (both high-quality HCP and clinical dataset), we compared the performance of our dictionary-based learning prior NODDI (DLpN) with the original NODDI for both single-shell and multi-shell data. Our results suggest that DLpN-derived NDI and ODI parameters for single-shell protocols are comparable with original multi-shell NODDI, and the protocol with b = 2000 s/mm performs the best (error ~ 5% in white and gray matter). This may allow NODDI evaluation of studies on single-shell data by multi-shell scanning of two subjects for DictNet f training.
神经突方向分散和密度成像（NODDI）可从多壳层扩散 MRI 数据评估细胞内、细胞外和自由水信号。这是一种深入了解脑组织微观结构的方法。在以前的研究中，由于拟合失败，特别是对于神经突密度指数（NDI），不鼓励对 NODDI 参数进行单壳重建。在这里，我们使用各向同性体积分数（f）作为先验，研究了使用单壳层数据创建稳健的 NODDI 参数图的可能性。先验估计是使用字典学习方法独立于 NODDI 模型约束进行的。首先，我们使用基于随机稀疏字典的网络（DictNet），该网络使用从体内和模拟扩散 MRI 数据获得的数据进行训练，来预测 f。在单壳层情况下，将平均扩散系数和无扩散加权（S）的原始 T 信号纳入字典中以进行 f 估计。然后，使用已知的 f 来使用 NODDI 框架估计 NDI 和方向分散指数（ODI）。将我们的模型中估计的 f 与模拟中的其他 f 估计器进行比较。此外，使用合成数据模拟和在 3T 扫描仪上收集的人类数据（高质量 HCP 和临床数据集），我们比较了基于字典的学习先验 NODDI（DLpN）与原始 NODDI 对单壳层和多壳层数据的性能。我们的结果表明，单壳层协议中基于 DLpN 的 NDI 和 ODI 参数与原始多壳层 NODDI 相当，并且 b=2000 s/mm 的协议性能最佳（白质和灰质的误差约为 5%）。这可能允许通过对两个对象进行多壳层扫描来进行 DictNet f 训练，从而对单壳层数据的 NODDI 评估进行研究。
