Low b-value Diffusion Tensor Imaging to Analyze the Dynamics of Cerebrospinal Fluid: Resolving Intravoxel Pseudorandom Motion into Ordered and Disordered Motions.

Abstract

Purpose: Analysis of cerebrospinal fluid (CSF) dynamics may be beneficial for understanding the mechanisms and diagnosis of several neurological diseases. Low b-value diffusion tensor imaging (low-b DTI) is useful for observing the slow and complex motion of the CSF. Theoretically, a mathematical framework suggests that low-b DTI provides the variance of the pseudorandom motion of the CSF. Furthermore, low-b DTI could provide comprehensive information on fluid dynamics. Accordingly, we proposed an analysis technique that resolves intravoxel pseudorandom motion into ordered (linear) and disordered (random) motions based on the mathematical framework.

Methods: The proposed analysis technique helps measure low-b DTI with multiple diffusion times and linearly fits its mean diffusivity (MD) with the diffusion time to obtain two parameters, double-slope V_v and y-intersect D_r, which represent the variance of the velocity distribution of linear motion and the diffusion coefficient of random motion, respectively. Seven healthy subjects were scanned to evaluate the proposed technique and investigate fluid dynamics in several representative ROIs.

Results: The obtained data showed the validity of the technique, repeatability, and consistency across the subjects in ROIs, such as the lateral ventricle (LV), third ventricle (3V), fourth ventricle (4V), and Sylvian fissure (SF). The obtained parameters V_v and D_r highlighted different characteristics of fluid dynamics in the representative ROIs: low V_v and low D_r in the LV, high V_v and moderate D_r in the 3V, and moderate V_v and moderate D_r in the 4V and SF.

Conclusion: The proposed analysis technique will facilitate a comprehensive investigation of the complex dynamics of the CSF using resolved parameters representing ordered and disordered motions.