Dear Jesper,

Thanks so much for your helps.

I would ask one more question. I used mricron to calculate the bvecs. But the v1 images are distorted and the lines are not aligned with the fa images. Also bvecs are different between subjects. I thought the acquisition protocol should have been the same for all the subjects. Is mricron a reliable tool for bvecs calculation?



On Aug 30, 2017 12:34, "Jesper Andersson" <[log in to unmask]uk> wrote:
Dear Arman,

Thanks so much for your response.

As Helen had also an EPI set with just one PE diresction I thought maybe she had found a way to resolve the problem of topup for just one PE direction. So how had she done that?

In her case she had one pair of images acquired with different PE-directions from which she could calculate the field with topup. She then applied that to a “full” data set that was acquired with only one PE direction. This is a fairly typical way of doing it. It is more unusual that people acquire the “full” data set with different PE-directions.

What does she mean by blipup/blipdown images? I thought they should have been the first b=0 volumes of different EPI sets with different PE directions.

Those are simply shorter ways of saying “image acquired with positive PE-blips” (blipup) and “image acquired with negative PE-blips” (blipdown).

Among the data which have been delivered to me I have an EPI set with 35 gradient directions which I don't know how to determine the PE direction. Besides I have also two phase images and two magnitude images acquired with two different echo times.

The two phase and magnitude images that you have is an alternative way of estimating the susceptibility field. That a look at the FSL help pages for information about what they are and how to use the FSL tools to transform these into a useful fieldmap.

Once you have that you can use the --field option in eddy to pass that fieldmap into eddy for doing the eddy current and motion correction. In order to do that you need to make sure that you find out the PE-direction of your data. This should be fairly easy by comparing the images to the example images you see here

The difficulty in your case will be to work out the “total readout time”. In order to do that you will need some information about the sequence that was used. Chances are that you need to wait for your colleagues to return from vacation to find that out.

Another thing you need to take care of when using a “traditional” fieldmap with eddy is to make sure it is accurately registered to the first volume in the data set that you feed into eddy.


Would you please help me to find how I can use these informations to correct my EPI for distortion and motion? Your help is highly appreciated. Everyone here is on holiday and I am left alone.