Diffusion spectrum imaging data for MSMT-CSD

1

Dear MRtrix3 Community,

I am working with diffusion MRI data acquired using a diffusion spectrum imaging (DSI) sequence, 120 directions with a maximum b-value of 6,800 s/mm². We were able to preprocess the data successfully, and overall the data quality appears very high.

However, I am unsure how best to proceed with MRtrix3 applications that require shelled acquisition schemes, such as MSMT-CSD, as I receive the error ‘[ERROR] DWI volumes could not be classified into b-value shells; gradient encoding may not represent a HARDI sequence‘. Is there a recommended workaround for using DSI data with such methods, or are there alternative approaches that would be more appropriate?

Any insights or recommendations would be much appreciated!

Best, Anna

2

Hi @anna_de, our group often collects DSI as well. We typically preprocess the data and then fit a 3dshore or MAPMRI model and use it to extrapolate a shelled scheme. This has worked very well in the past. We usually get a maximum b value of 5000 mm/s2 and extrapolate to the HCP-YA sampling scheme.

1 Like