2021/02/02 by Abrar Faiyaz, Faiyaz, Abrar, Marvin M. Doyley +7
Medicine · #Advanced MRI Techniques and Applications #Advanced Neuroimaging Techniques and Applications #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #FOS: Electrical engineering #FOS: Physical sciences #Image and Video Processing (eess.IV) #MRI in cancer diagnosis #Machine Learning (cs.LG) #Medical Physics (physics.med-ph) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2102.02772
openalex publication_date 2021/02/02 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
Neurite orientation dispersion and density imaging (NODDI) enables the\nassessment of intracellular, extracellular and free water signals from\nmulti-shell diffusion MRI data. It is an insightful approach to characterize\nbrain tissue microstructure. Single-shell reconstruction for NODDI parameters\nhas been discouraged in previous studies caused by failure when fitting,\nespecially for the neurite density index (NDI). Here, we investigated the\npossibility of creating robust NODDI parameter maps with single-shell data,\nusing the isotropic volume fraction (fISO) as prior. Prior estimation was made\nindependent of the NODDI model constraint using a dictionary learning approach.\nFirst, we used a stochastic sparse dictionary-based network (DictNet) in\npredicting fISO which is trained with data obtained from in vivo and simulated\ndiffusion MRI data. In single-shell cases, the mean diffusivity (MD) and raw T2\nsignal with no diffusion weighting (S0) was incorporated in the dictionary for\nthe fISO estimation. Then, the NODDI framework was used with the known fISO to\nestimate the NDI and orientation dispersion index (ODI). The fISO estimated by\nour model was compared with other fISO estimators in the simulation. Further,\nusing both synthetic data simulation and human data collected on a 3T scanner,\nwe compared the performance of our dictionary-based learning prior NODDI (DLpN)\nwith the original NODDI for both single-shell and multi-shell data. Our results\nsuggest that DLpN derived NDI and ODI parameters for single-shell protocols are\ncomparable with original multi-shell NODDI, and protocol with b=2000 s/mm2\nperforms the best (error ~5% in white and grey matter). This may allow NODDI\nevaluation of studies on single-shell data by multi-shell scanning of two\nsubjects for DictNet fISO training.\n