Single‐shell NODDI using dictionary‐learner‐estimated isotropic volume fraction

Abstract

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. Singleshell 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 ISO ) 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 ISO . In single-shell cases, the mean diffusivity and raw T 2 signal with no diffusion weighting (S 0 ) was incorporated in the dictionary for the f ISO estimation. Then, the NODDI framework was used with the known f ISO to estimate the NDI and orientation dispersion index (ODI). The f ISO estimated using our model was compared with other f ISO 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 2 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 ISO training.

Potential Pitfalls of Using Fractional Anisotropy, Axial Diffusivity, and Radial Diffusivity as Biomarkers of Cerebral White Matter Microstructure

“…water molecules within our brains. For example, dMRI data has been used to: (1) derive several different quantitative measures [e.g., fractional anisotropy, axial diffusivity, radial diffusivity, mean diffusivity (Beaulieu, 2002;Alexander et al, 2019); axial kurtosis, radial kurtosis, mean kurtosis, maximum directional kurtosis, axonal water fraction (Fieremans et al, 2011;Henriques et al, 2021); neurite orientation dispersion, neurite density index, isotropic volume fraction (Zhang et al, 2012;Faiyaz et al, 2021); etc.] that reflect slightly different aspects of tissue microstructure, (2) non-invasively map the brain's white matter pathways using deterministic (Mori et al, 1999) and/or probabilistic (Behrens et al, 2003) tractography approaches (Maier-Hein et al, 2017;Jeurissen et al, 2019), and (3) indirectly measure brain function (Le Bihan et al, 2006b;Le Bihan, 2007;Abe et al, 2017).…”

Artificial intelligence for diffusion MRI-based tissue microstructure estimation in the human brain: an overview

“…The following approachesnumerical analysis, Monte Carlo, animal models, phantoms, tissue fixation experiments, etc.-are commonly used to validate these models (50). The validation modes can often lead to good training data sources in DL/ML practice (30,51).…”

“…. AI in dMRI microstructure estimation AI algorithms have been applied widely throughout the dMRI field, including signal reconstruction, denoising, detection and removal of artifacts, segmentation, co-registration, spatial and angular super-resolution of the dMRI signal, and tissue microstructure modeling (18,21,24,30,(52)(53)(54). Table 1 identifies such approaches (Block B-D) and lists relevant biophysical models on which DL/ML approaches have been used (Block A).…”

“…This can be easily termed "prior regularization" since it pushes the solution out of local minima. The objective values have been shown to improve and reduce bias when the starting points of the MLE are determined through an adapted multiple-layer perceptron (MLP) (30,58).…”

“…AI algorithms have evolved with DL/ML techniques due to skyrocketing computing power and resources (13,14). Recently, AI-based approaches have been proposed to address issues such as denoising and artifact reduction (15-18), fiber tractography (19-21), resolution enhancement (22)(23)(24)(25), and quantification of microstructural properties (20)(21)(22)(26)(27)(28)(29)(30)(31). These techniques have been particularly useful for working with clinical and challenging datasets and have led to advances in dMRI parameter mapping and image quality assessment and improvement.…”