Advances in computational and statistical diffusion MRI

Abstract

Computational methods are crucial for the analysis of diffusion magnetic resonance imaging (MRI) of the brain. Computational diffusion MRI can provide rich information at many size scales, including local microstructure measures such as diffusion anisotropies or apparent axon diameters, whole-brain connectivity information that describes the brain's wiring diagram and population-based studies in health and disease. Many of the diffusion MRI analyses performed today were not possible five, ten or twenty years ago, due to the requirements for large amounts of computer memory or processor time. In addition, mathematical frameworks had to be developed or adapted from other fields to create new ways to analyze diffusion MRI data. The purpose of this review is to highlight recent computational and statistical advances in diffusion MRI and to put these advances into context by comparison with the more traditional computational methods that are in popular clinical and scientific use. We aim to provide a high-level overview of interest to diffusion MRI researchers, with a more in-depth treatment to illustrate selected computational advances.

The role of diffusion MRI in neuroscience

“…43,45 Further studies have shown that individual differences in measures of local white matter microstructure correlate with variations in physiological properties of the fibre pathways. [46][47][48][49][50][51][52] For example, paired-pulse transcranial magnetic stimulation can be used to probe the functional connectivity of a cortico-cortical connection. A conditioning pulse applied to the dorsal premotor cortex of one hemisphere will modulate the excitability of the primary motor cortex in the other hemisphere; the degree of modulation can be used to estimate the functional connectivity between the two cortical areas.…”

Reducing variability in along-tract analysis with diffusion profile realignment

“…O'Donnell et al (2009) previously noticed the potential issue introduced by misalignment between subjects mentioning that "improved cross-subject alignment is of interest [...] as the high-frequency variations seen in individual subjects [...] are smoothed in the group average". While many methods for registering dMRI volumes or streamlines were developed (see, e.g., O'Donnell et al (2017) for a review), they do not directly address the issue of possible residual misalignment between the end points after extracting the representative streamlines of each subject. To ensure an adequate comparison between subjects, one must make sure that each streamline corresponds to the same underlying anatomical location.…”

Fine-Grained Topography and Modularity of the Macaque Frontal Pole Cortex Revealed by Anatomical Connectivity Profiles

“…invasive tract-tracing techniques, but DTI has proven to be an indispensable method and can offer invaluable insights for neuroscience [88] and neuroanatomy [89,90], including the discovery of new pathways [31], the description of whole-brain connectivity information [32], and the refinement of brain regions [28].…”

“…In view of the importance of the diversity of functions of the monkey FPC and the lack of detailed anatomical connection information in comparison with the previous tracer results [19,26,27], as well as to pave the way for a systematic follow-up study using tracer injections, a study of the topological organization properties of the macaque FPC is necessary and attractive. Recently, connectivitybased parcellation (CBP) has been a powerful framework for mapping the human brain [28][29][30] and may provide a better picture of regional parcellation and anatomical connectivity information [31,32] as well as allowing the target areas of tracer injections to be chosen less blindly. In this study, we provided a tractography-based parcellation scheme that applied a machine-learning algorithm to obtain a fine-grained subdivisions of the macaque FPC, and then revealed their subregional connections.…”