Dear Alberto,
it is pretty much impossible to say with certainty that a given set of differences would not have an effect on the accuracy of the fieldmap estimation. In the end it will have to be a judgement call based on looking at the estimated field and at the pair of corrected images. You want the field (--fout) to be smooth, look like a susceptibility field and to be free of any obvious artefacts. For the corrected images (--iout) you can load them into fslview or fsleyes and run a movie loop. You want to see as little obvious frame to frame differences as possible.
In an ideal world we would never make any mistakes, and would also acquire a pair of images with opposing PE direction (admittedly a rather limited view of an ideal world). But when we make mistakes we end up with these judgement calls. I have myself been involved in running topup with one EPI and a T2 weighted image (infinite bandwidth), and the results have looked good.
Best regards Jesper
On 22/08/2022, 15:05, "FSL - FMRIB's Software Library on behalf of Alberto Inuggi" <[log in to unmask] on behalf of [log in to unmask]> wrote:
Dear all
We realized that our posterior-anterior (PA) epi sequence for susceptibility correction is not an exact copy (but for the PE direction) of the AP epi we want to correct.
these are the different parameters:
GOOD_PA sequence (exact copy of the EPI_AP epi we want to correct, with PE direction inverted)
"EchoTime": 0.033,
"RepetitionTime": 0.72,
"FlipAngle": 52,
"RefLinesPE": 12,
"MultibandAccelerationFactor": 8,
"EchoTrainLength": 90,
"BandwidthPerPixelPhaseEncode": 19.157,
"EffectiveEchoSpacing": 0.000580003,
"DerivedVendorReportedEchoSpacing": 0.000580003,
"TotalReadoutTime": 0.0516202,
BAD_PA sequence
"EchoTime": 0.058,
"RepetitionTime": 7.06,
"FlipAngle": 90,
"RefLinesPE": 24,
"EchoTrainLength": 45,
"BandwidthPerPixelPhaseEncode": 42.735,
"ParallelReductionFactorInPlane": 2,
"EffectiveEchoSpacing": 0.00026,
"DerivedVendorReportedEchoSpacing": 0.000520001,
"TotalReadoutTime": 0.02314
I applied topup with both PA versions (setting the two proper TotalReadOutTime).
After visual inspection, they seem to similarly and succesfully correct for the A->P stretching. but since they are very different sequences (e.g. EPI_AP and GOOD_PA are multiband, BAD_PA not and different TR/ET ) I'm not sure I can do it .
do you have any suggestion?
thanks in advance
Alberto
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=FSL&A=1
This message was issued to members of www.jiscmail.ac.uk/FSL, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=FSL&A=1
This message was issued to members of www.jiscmail.ac.uk/FSL, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
|