 |
 |
Dear Jesper It works !!! As you suggested, the problem came from the last volume which was added by the Philips MRI (a sort of average of all volumes) once i removed it, all worked well Thank you for your help Christophe ________________________________________ De : FSL - FMRIB's Software Library [[log in to unmask]] de la part de Maggia, Christophe [[log in to unmask]] Envoyé : lundi 2 mars 2015 14:02 À : [log in to unmask] Objet : Re: [FSL] Problem with eddy Dear Jesper I obtained these bvec and bval directly from the nifti converter. Do you mean that there should be no vectors equal to zero ? even the first one (B0=0) where there is no gradient ? Can i put it to 1 for example ? For the last one, i'm going to try without that last volume because it does not represent a direction in the DTI. I will tell you if it works Thanks Christophe ________________________________________ De : FSL - FMRIB's Software Library [[log in to unmask]] de la part de Jesper Andersson [[log in to unmask]] Envoyé : samedi 28 février 2015 11:08 À : [log in to unmask] Objet : Re: [FSL] Problem with eddy Dear Christophe, I notice you have two bvecs in MAP-VS-287_DTI_32_dirCM.bvec with zero length (the first and the last one). The first on has a zero bval so you can set that to anything with a unity length, the last one has a bval of 1000 so you should find out what it really should be. Try that and let me know if that fixes your eddy problem. Jesper On 26 Feb 2015, at 17:57, Maggia, Christophe <[log in to unmask]> wrote: > Dear Jesper > > I uploaded all the files (eddyCM.zip) > and this is the command line i ran > > eddy --imain=MAP-VS-287_DTI_32_dir_SENSECM.nii --mask=MAP-VS-287_DTI_32_dir_SENSE_brain_maskCM.nii --index=indexCM.txt --acqp=acqparamsCM.txt --bvecs=MAP-VS-287_DTI_32_dirCM.bvec --bvals=MAP-VS-287_DTI_32_dirCM.bval --fwhm=0 --topup=topup_resultCM --flm=quadratic --out=DTI_sec.nii > > Thank you again > > Christophe > > ________________________________ > De : FSL - FMRIB's Software Library [[log in to unmask]] de la part de Jesper Andersson [[log in to unmask]] > Envoyé : jeudi 26 février 2015 17:35 > À : [log in to unmask] > Objet : Re: [FSL] Problem with eddy > > Dear Christophe, > > I had a look at your topup —fout output, and that looks fine (apart from being sign swapped, but that doesn’t really matter). > > Can you please upload (to the same place as before) all files referred to on your eddy command line along with the exact command you ran? > > Jesper > > On 26 Feb 2015, at 11:26, Maggia, Christophe <[log in to unmask]<mailto:[log in to unmask]>> wrote: > > Thank you again Jesper > > I uploaded it (fieldmapCMaggia.nii) > I calculated the readout time with the recommendations of Philips, so it should be fine > > I don't know if you have seen my previous message but i tried to use the output file with applytopup and i also have an error : > ApplyTopup : : msg=Invalid combination of phase-encode vectors for least-square restoration > > maybe the two are linked but i still can't find a solution... > > Christophe > ________________________________ > De : FSL - FMRIB's Software Library [[log in to unmask]<mailto:[log in to unmask]>] de la part de Jesper Andersson [[log in to unmask]<mailto:[log in to unmask]>] > Envoyé : jeudi 26 février 2015 11:46 > À : [log in to unmask]<mailto:[log in to unmask]> > Objet : Re: [FSL] Problem with eddy > > Dear Christophe, > > Indeed, i used the b02b0.cnf configuration file (it is included in FSL right ?, i don't need to download it ? because i just wrote the file name) > > that should be sufficient. > > The output from topup looks pretty good, (i don't have much experience with field map but i don't see any problem) > the minimum value is -245.7 and the maximum is 289.7 > > these are quite high and might be indicative of an error. Though if you just guessed the readout time they might still be kosher. > > i see no problem on the edge of the FOV > i can join the fout file if you want > > Yes, can you please upload it to > > https://oxfile.ox.ac.uk/oxfile/work/extBox?id=72139C7E463068D5F > > Jesper > > Thanks again for your help > > Christophe > ________________________________________ > De : FSL - FMRIB's Software Library [[log in to unmask]<mailto:[log in to unmask]><mailto:[log in to unmask]>] de la part de Jesper Andersson [[log in to unmask]<mailto:[log in to unmask]><mailto:[log in to unmask]>] > Envoyé : mercredi 25 février 2015 11:23 > À : [log in to unmask]<mailto:[log in to unmask]><mailto:[log in to unmask]> > Objet : Re: [FSL] Problem with eddy > > Dear Christophe, > > i have seen something similar to this. That time it was because topup hadn’t converged properly, which was in turn caused by not using the recommended configuration file (b02b0.cnf). Can you please confirm that it was used, and also that the output from topup looks good. Look in particular for very large values at the edge of the FOV in the —fout output. > > Jesper > > > On 25 Feb 2015, at 09:26, Christophe Maggia <[log in to unmask]<mailto:[log in to unmask]><mailto:[log in to unmask]>> wrote: > > Hello everyone > > I'm trying to use eddy after topup to correct susceptibility and eddy current distorsions on DTI images (with 32 directions) > topup is working very well but when i want to use eddy i get a weird error. > At first, i thought it was a memory problem, but i tested it with another computer with the same results. > > my command line is : > eddy --imain=MAP-VS-287_DTI_32_dir_SENSEmcverter.nii --mask=b0_Pbrain_mask.nii.gz --index=index.txt --acqp=acqparams.txt --bvecs=MAP-VS-287_DTI_32_dir.bvec --bvals=MAP-VS-287_DTI_32_dir.bval --fwhm=0 --topup=my_topup_results --flm=quadratic --out=DTI_ec.nii > > and i get this > > Running Register > Segmentation violation, Address not mapped, Offending address = 0x512ae0000 > eddy 0x000000010010ad86 SPLINTERPOLATOR::Splinterpolator<float>::value_at(double const*) const > eddy 0x00000001000c77d0 NEWIMAGE::volume<float>::splineinterpolate(float, float, float) const > eddy 0x00000001000e5f20 NEWIMAGE::volume<float>::interpolate(float, float, float) const > eddy 0x000000010002a084 void NEWIMAGE::raw_general_transform<float>(NEWIMAGE::volume<float> const&, NEWMAT::Matrix const&, NEWIMAGE::volume4D<float> const&, std::vector<int, std::allocator<int> > const&, std::vector<int, std::allocator<int> > const&, NEWMAT::Matrix const*, NEWMAT::Matrix const*, NEWIMAGE::volume<float>&, NEWIMAGE::volume4D<float>&, NEWIMAGE::volume<char>*) > eddy 0x000000010002b3d4 void NEWIMAGE::affine_transform<float>(NEWIMAGE::volume<float> const&, NEWMAT::Matrix const&, NEWIMAGE::volume<float>&, NEWIMAGE::volume<char>&) > eddy 0x0000000100019269 EDDY::ECScan::field_for_model_to_scan_transform(boost::shared_ptr<NEWIMAGE::volume<float> const>, NEWIMAGE::volume<float>*, NEWIMAGE::volume<float>*) const > eddy 0x0000000100030f2c EDDY::EddyUtils::transform_model_to_scan_space(NEWIMAGE::volume<float> const&, EDDY::ECScan const&, boost::shared_ptr<NEWIMAGE::volume<float> const>, bool, NEWIMAGE::volume<float>&, NEWIMAGE::volume<float>*, NEWIMAGE::volume4D<float>*) > eddy 0x0000000100036044 EDDY::EddyUtils::param_update(NEWIMAGE::volume<float> const&, boost::shared_ptr<NEWIMAGE::volume<float> const>, NEWIMAGE::volume<float> const&, EDDY::Parameters, bool, EDDY::ECScan&, NEWMAT::ColumnVector*) > eddy 0x0000000100004cfc Register(EDDY::EddyCommandLineOptions const&, EDDY::ScanType, EDDY::ECScanManager&, unsigned int, NEWMAT::Matrix&, NEWMAT::Matrix&) > eddy 0x0000000100005e7d main > eddy 0x0000000100001ca8 start > ??? 0x000000000000000b 0x0 > > > The "Offending address" is almost the same everytime 0x512****** > Do anyone know what i can do to make it work ? Please