ENH: MRIQC for DWI

[oesteban]

CC/ @mckenziephagen @arokem @teresamg

[oesteban]

Interesting addition of an estimator of the number of shells (nipreps/dmriprep#64):

95497ca#diff-0afea89f34b27abdc8fb70e57182420cdfb40a7779a3ca808feebc2978e12e56R145-R158

CC/ @edickie @arokem @mckenziephagen

I have tested with my data and it seems to work pretty well. The idea is that the estimated number of shells is very very high (e.g., 8 or 9), this is probably a DSI dataset.

To test it:

import numpy as np
from mriqc.interfaces.diffusion import NumberOfShells

bvals =  np.loadtxt("path/to/dwi.bval")

print(NumberOfShells(in_data=bvals.tolist()).run().outputs.n_shells)

let me know what you think