Studies have demonstrated that many regions in the human brain include multidirectional fiber tracts, in which the diffusion of water molecules within image voxels does not follow a Gaussian distribution. Therefore, the conventional diffusion tensor imaging (DTI) that hypothesizes a single fiber orientation within a voxel is intrinsically incapable of revealing the complex microstructures of brain tissues. Diffusion spectrum imaging (DSI) employs a pulse sequence with different b-values along multiple gradient directions to sample the diffusion information of water molecules in the entire q-space and then quantitatively estimates the diffusion profile using a probability density function with a high angular resolution. Studies have suggested that DSI can reliably observe the multidirectional fibers within each voxel and allow fiber tracking along different directions, which can improve fiber reconstruction reflecting the true but complicated brain structures that were not observed in the previous DTI studies. Moreover, with increasing angular resolution, DSI is able to reveal new neuroimaging biomarkers used for disease diagnosis and the prediction of disorder progression. However, so far, this method has not been used widely in clinical studies, due to its overly long scanning time and difficult post-processing. Within this context, the current paper aims to conduct a comprehensive review of DSI research, including the fundamental principles, methodology, and application progress of DSI tractography. By summarizing the DSI studies in recent years, we propose potential solutions towards the existing problem in the methodology and applications of DSI technology as follows: (1) using compressed sensing to undersample data and to reconstruct the diffusion signal may be an efficient and promising method for reducing scanning time; (2) the probability density function includes more information than the orientation distribution function, and it should be extended in application studies; and (3) large-sample study is encouraged to confirm the reliability and reproducibility of findings in clinical diseases. These findings may help deepen the understanding of the DSI method and promote its development in clinical applications.
11 November 2022 in Frontiers in Human Neuroscience
Background: Previous studies demonstrated that cerebellar subregions are involved in different functions. Especially the cerebellar anterior lobe (CAL) and cerebellar posterior lobe (CPL) have been postulated to primarily account for sensorimotor and cognitive function, respectively. However, the functional connectivity (FC) alterations of CAL and CPL, and their relationships with behavior performance in chronic stroke participants are unclear so far.Materials and methods: The present study collected resting-state fMRI data from thirty-six subcortical chronic stroke participants and thirty-eight well-matched healthy controls (HCs). We performed the FC analysis with bilateral CAL and CPL as seeds for each participant. Then, we detected the FC difference between the two groups by using a two-sample t-test and evaluated the relationship between the FC and scores of motor and cognitive assessments across all post-stroke participants by using partial correlation analysis.Results: The CAL showed increased FCs in the prefrontal cortex, superior/inferior temporal gyrus, and lingual gyrus, while the CPL showed increased FCs in the inferior parietal lobule, precuneus, and cingulum gyrus in the stroke participants compared with HCs. Moreover, the FC alteration in the right CAL and the right CPL were negatively correlated with executive and memory functions across stroke participants, respectively.Conclusion: These findings shed light on the different increased FC alteration patterns of CAL and CPL that help understand the neuro-mechanisms underlying behavior performance in chronic stroke survivors.