Motion related parameters

Dear Mrtrix experts

I was wondering if there is a way to get motion parameters in order to see if they lie within the accepted range.
I used dwipreproc script for preprocessing of DWIs.

Thanks in advance.

Hi Hamed,

The easiest way to get access to these data is to run the dwipreproc script with the -nocleanup option. Inside the temporary working directory of the script, you’ll find the text file generated by eddy that contains the motion parameters it estimated.


Hi Rob
Thanks for the reply.

Could you please specify which file you mean. Do you mean dwi_post_eddy (Eddy_MOVEMENT_RMS) file or something else?

And what criteria would you suggest to exclude subjects?


The dwipreproc script is essentially a convenience wrapper for FSL’s topup and eddy tools, so I don’t know the precise nuances of the additional files output by these commands. I’m pretty sure that the file dwi_post_eddy.eddy_parameters will contain per-volume translation / rotation / eddy current estimates, but you’d have to consult the FSL documentation to find out the precise formatting and axis conventions of the file. The dwi_post_eddy.eddy_movement_rms appears to have 2 values per volume; I’d guess these would be something like root-mean-square of the 3 translation and rotation components, but it would be a guess. You’d have to contact the guys at FMRIB to find out for sure.

I’ve never worked on large cohorts where I’ve had to deal with subject exclusion, so I honestly have no idea what criteria would be appropriate here. You may be able to find examples out in the literature who have quoted their chosen thresholds.


Thanks Rob for the help.