Quantitative scalar measures of diffusion MRI datasets are subject to normal variability across subjects, but potentially abnormal values may yield essential information to support analysis of controls and patients cohorts. However, small changes in the measured signal due to differences in scanner hardware or reconstruction methods in parallel MRI may translate into small differences in diffusion metrics such as fractional anisotropy (FA) and mean diffusivity (MD). In the presence of disease, these small variations are entangled in the genuine biological variability between subjects. In this work,we propose a new harmonization algorithm based on adaptive dictionary learning to mitigate the unwanted variability caused by different scanner hardware while preserving the natural biological variability present in the data.
Samuel St-Jean, Max A. Viergever and Alexander Leemans Harmonization of diffusion MRI datasets using automated feature extraction from multiple scanners Proceedings of: ISMRM 2020 Benelux chapter, 2020
Accepted for an oral presentation.