Steventon, J. J, Trueman, R. C, Rosser, A. E & Jones, DK. (2016). Robust MR-based approaches to quantifying white matter structure and structure/function alterations in Huntington's disease. Journal of Neuroscience Methods,265 2-12. Netherlands: Elsevier. Retrieved from https://doi.org/10.1016/j.jneumeth.2015.08.027
Background: Huge advances have been made in understanding and addressing confounds in diffusion MRI data to quantify white matter microstructure. However, there has been a lag in applying these advances in clinical research. Some confounds are more pronounced in HD which impedes data quality and interpretability of patient-control differences. This study presents an optimised analysis pipeline and addresses specific confounds in a HD patient cohort. Method: 15 HD gene-positive and 13 matched control participants were scanned on a 3T MRI system with two diffusion MRI sequences. An optimised post processing pipeline included motion, eddy current and EPI correction, rotation of the B matrix, free water elimination ( FWE ) and tractography analysis using an algorithm capable of reconstructing crossing fibres. The corpus callosum was examined using both a region-of-interest and a deterministic tractography approach, using both conventional diffusion tensor imaging ( DTI )-based and spherical deconvolution analyses. Results: Correcting for CSF contamination significantly altered microstructural metrics and the detection of group differences. Reconstructing the corpus callosum using spherical deconvolution produced a more complete reconstruction with greater sensitivity to group differences, compared to DTI-based tractography. Tissue volume fraction ( TVF ) was reduced in HD participants and was more sensitive to disease burden compared to DTI metrics. Conclusion: Addressing confounds in diffusion MR data results in more valid, anatomically faithful white matter tract reconstructions with reduced within-group variance. TVF is recommended as a complementary metric, providing insight into the relationship with clinical symptoms in HD not fully captured by conventional DTI metrics.
Open Access Journal Article