Interpretation of modulated NODDI maps

[fkm24]

Hi AMICO developers,

I have just started using AMICO to fit the NODDI models on our DWI datasets and the results are very encouraging. I've noticed an option to generate modulated ODI and NDI maps using the tissue fraction (inverted free water maps), and I have a query regarding their interpretation for subsequent voxel-wise or ROI comparisons between groups. In the event of significant group differences, would it be valid to assert that the observed ODI/NDI differences exist even after accounting for atrophy/CSF?

Many thanks for your help!

Best Wishes,
Elijah