Dear Matthias,
can you please check that the dimensions (matrix and voxel sizes) of your mask is identical to that of your data. If that isn’t the problem I am afraid that I will not be able to look at your data until after the release.
Best regards Jesper
> On 19 Dec 2018, at 10:40, Matthias Heyne <[log in to unmask]> wrote:
>
> Dear experts,
>
> I've been able to run eddy_cuda on various data sets via our computing cluster at Boston University but can't get the tool to work on a GE data set containing diffusion volumes with only a single b0 (+ 25 b=1000 volumes). The piped output below (using the very verbose flag) from running the tool without slice-to-volume correction indicates that the tool seems to be recognizing only 25 volumes within the nifti input file. I'm suspecting that the first diffusion direction (-1, 0, 0) in the bvecs supplied below might be throwing the tool off?
> I've also observed very similar behavior with a data set collected on a Siemens scanner containing only a single b0 at the beginning of the diffusion volume.
>
> Best,
> Matthias Heyne
> Postdoctoral Associate, Speech Lab at Boston University (PI Frank Guenther)
>
> bvecs: (rotated 90° from bvecs file supplied to eddy_cuda)
> 0 0 0
> -1 0 0
> -0.849 0.528 0
> 0.108 0.565 0.818
> -0.884 -0.345 -0.315
> 0.003 -0.736 0.677
> 0.868 -0.238 0.436
> -0.799 0.37 0.475
> 0.162 0.987 0
> -0.866 -0.129 0.483
> 0.212 -0.936 0.281
> -0.068 -0.892 -0.446
> -0.55 -0.544 -0.634
> 0.435 -0.422 0.795
> 0.599 0.78 0.182
> 0.525 0.03 -0.851
> 0.6 -0.688 0.409
> -0.653 -0.06 -0.755
> -0.207 -0.076 -0.975
> 0.413 -0.699 -0.584
> 0.436 0.822 -0.366
> -0.462 0.874 0.148
> 0.503 0.488 -0.713
> -0.824 -0.53 0.202
> -0.297 0.349 0.889
> 0.04 0.318 -0.947
>
>
> qlog content:
> terminate called after throwing an instance of 'EDDY::EddyException'
> what(): eddy: msg=CudaVolume::operator*=: Mismatched volumes
> /var/spool/sge/scc-ec2/job_scripts/60062: line 14: 4802 Aborted eddy_cuda --imain=../../DWI_vol_for_eddy_cuda_denoised_degibbsed.nii.gz --mask=DWI_mask.nii --acqp=../../../acqp_DWI_volumes_testing.txt --index=../../index.txt --bvecs=/projectnb/busplab/Experiments/CWS/subcort_parc/preprocessed_data/SF109_2/orig_bvecs --bvals=/projectnb/busplab/Experiments/CWS/subcort_parc/preprocessed_data/SF109_2/orig_bvals --niter=5 --fwhm=2,0,0,0,0 --repol --cnr_maps --residuals --out=eddy_cuda_pre-processed_DWI_no_susceptibility_correction_w_denoising_and_degibbsing --very_verbose > eddy_cuda_log_no_susceptibility_correction_w_denoising_and_degibbsing.txt
>
> very verbose piped output:
> Reading images
> Performing volume-to-volume registration
> Running Register
>
> ...................Allocated GPU # 0...................
> Loading prediction maker
> Scan: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
> Evaluating prediction maker model
> Initial guess for hyperparameters: 2.222659
> 0.450000
> 2.469621
>
> cf = 24461.24746
> cf = 24461.24746
> p = 3.22266 0.45000 2.46962
> cf = 24461.24746
> p = 3.22266 0.45000 2.46962
> cf = 24461.24746
> p = 3.22266 0.45000 2.46962
> cf = 24461.24746
> p = 3.22266 0.45000 2.46962
> cf = 24461.24746
> p = 3.22266 0.45000 2.46962
> cf = 24461.24746
> p = 3.22266 0.45000 2.46962
> cf = 24461.24746
> p = 3.22266 0.45000 2.46962
> cf = 24461.24746
> p = 3.22266 0.45000 2.46962
> cf = 24461.24746
> p = 3.22266 0.45000 2.46962
> cf = 24461.24746
> p = 3.22266 0.45000 2.46962
> cf = 24441.18466
> p = 3.22198 0.39223 2.22218
> cf = 24435.37586
> p = 3.34079 0.42215 2.19777
> cf = 24435.37586
> p = 3.34079 0.42215 2.19777
> cf = 24417.50838
> p = 3.27330 0.42332 2.38483
> cf = 24417.50838
> p = 3.27330 0.42332 2.38483
> cf = 24417.50838
> p = 3.27330 0.42332 2.38483
> cf = 24417.50838
> p = 3.27330 0.42332 2.38483
> cf = 24417.50838
> p = 3.27330 0.42332 2.38483
> cf = 24416.79606
> p = 3.23403 0.41228 2.31355
> cf = 24416.79606
> p = 3.23403 0.41228 2.31355
> cf = 24416.79606
> p = 3.23403 0.41228 2.31355
> cf = 24416.79606
> p = 3.23403 0.41228 2.31355
> cf = 24416.38565
> p = 3.25087 0.41897 2.36616
> cf = 24416.32718
> p = 3.24159 0.41424 2.36955
> cf = 24416.27318
> p = 3.21839 0.41772 2.33716
> cf = 24416.0885
> p = 3.23549 0.41463 2.33559
> cf = 24416.0885
> p = 3.23549 0.41463 2.33559
> cf = 24416.0885
> p = 3.23549 0.41463 2.33559
> cf = 24416.08158
> p = 3.24743 0.41608 2.33564
> cf = 24416.04264
> p = 3.23814 0.41688 2.34307
> cf = 24416.04264
> p = 3.23814 0.41688 2.34307
> cf = 24416.02975
> p = 3.23720 0.41570 2.33549
> cf = 24416.02975
> p = 3.23720 0.41570 2.33549
> cf = 24416.02975
> p = 3.23720 0.41570 2.33549
> cf = 24416.01819
> p = 3.23997 0.41633 2.34146
> cf = 24416.01819
> p = 3.23997 0.41633 2.34146
> cf = 24416.01819
> p = 3.23997 0.41633 2.34146
> cf = 24416.01819
> p = 3.23997 0.41633 2.34146
> cf = 24416.01796
> p = 3.23644 0.41634 2.33509
> cf = 24416.01796
> p = 3.23644 0.41634 2.33509
> cf = 24416.01785
> p = 3.23717 0.41613 2.33706
> cf = 24416.01733
> p = 3.23823 0.41626 2.33898
> cf = 24416.01733
> p = 3.23823 0.41626 2.33898
> cf = 24416.01733
> p = 3.23823 0.41626 2.33898
> cf = 24416.01725
> p = 3.23726 0.41629 2.33776
> Fudging parameters
> Parameters start out as 3.23726
> 0.41629
> 2.33776
>
> Parameters end up as 3.23726
> 0.41629
> 4.64034
>
> Estimated hyperparameters: 3.23726
> 0.41629
> 4.64034
>
> Checking for outliers
> Making predictions for scans: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
> Checking scan: 0 eddy: msg=CudaVolume::operator*=: Mismatched volumes
> ########################################################################
>
> To unsubscribe from the FSL list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|