Preprocessing data acquired using CMRR protocol

AFNI Message Board
1

AFNI experts,

I used Multi-Band Accelerated EPI Pulse Sequences, so-called CMRR protocol.
TR = 460 ms, voxel size = 2.7 mm

When I used this protocol, additional 4 images for distortion corrections (two for each of polarity, A->P or P->A) were acquired before actual fMRI scanning.
Also, for each run, there were additional reference images.
I heard that I can use FSL function, “topup” for correcting distortion using the four images.
Also, I heard that for each run, I can use the reference image when I perform motion correction.

Is there any AFNI script for this procedure of preprocessing?
I don’t know how to use these additional images in addition to conventional AFNI preprocessing (guided by afni_proc.py)
Thank you in advance.

Sincerely,
Sungshin Kim

2

Hi Sungshin,

We refer to those 2 pairs of distortion correction images as forward and reverse blip datasets. The forward direction should match that of the functional EPI data, while the reverse is the opposite.

Those 2 datasets can be passed to afni_proc.py using -blip options:
-blip_forward_dset FORWARD+orig
-blip_reverse_dset REVERSE+orig \

The resulting proc script will have a ‘blip’ processing block, which you are welcomed to review.

There is no option to apply per-run reference images (except to pre-process using multiple afni_proc.py commands). If shimmimg was applied between runs, it would be a good idea to use per-run registration. Otherwise, it is unlikely to matter much.

  • rick