Mismatch between b-table and the loaded images

Software Support
1

Dear expert,

I am currently using QSIPrep for diffusion MRI data processing, and I encountered an issue that I would greatly appreciate your expertise on.

Issue Description:
During the preprocessing pipeline, QSIPrep fails to complete successfully due to the presence of NaN (Not a Number) values in the b-vector. Would this issue potentially stem from data acquisition (e.g., gradient coil errors) or a post-processing step in QSIPrep?Could you advise on the recommended method to handle NaN values in b-vectors within QSIPrep output? Is there a built-in function or preprocessing step to address this?

Raw data b_value:

1.28951 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928 999.928

Raw data b_vector：

0 0.59185 0.704632 0.53046 0.391319 0.415961 0.61739 0.76246 0.843651 0.729609 0.60166 0.439662 0.154649 0.24892 0.33509 0.10002 0.383919 0.590209 0.764123 0.795722 0.905201 0.915953 0.781519 0.601858 0.547549 0.36422 -0.00194101 -0.363009 -0.0705341 0.029174 -0.19403 -0.14595 0.14974 0.17442 0.304279 0.510337 0.691446 0.876011 0.934537 0.981965 0.928881 0.812628 0.612371 0.413002 0.379411 0.193601 -0.15864 -0.206249 -0.511619 -0.2352 -0.47899 -0.425582 -0.346371 -0.0931953 -0.172999 0.115089 0.189249 0.390969 0.573943 0.773103 0.923053 0.990846 0.986576 0.847758 0.925939 0.82071 0.868903 0.76527 0.492381 0.273341 0.205031 0.0709298 0.022953 -0.11441 -0.30048 -0.35309 -0.663644 -0.64413 -0.489471 -0.698512 -0.678087 -0.620732 -0.764409 -0.517301 -0.65165 -0.344031 0.013239 -0.1354 0.0568902 0.249239 0.433328 0.497631 0.625829 0.789101 0.904397 0.800287 0.950642
0 0.50068 0.607081 0.676441 0.530369 0.325681 0.30108 0.42688 0.505211 0.682239 0.771469 0.834784 0.972277 0.863618 0.708361 0.695581 0.0848807 0.0606569 -0.00784403 0.213161 0.30333 0.394911 0.592109 0.701218 0.829989 0.92995 0.997874 0.878328 0.950102 0.84594 0.648649 0.46526 0.515161 0.30363 -0.187799 -0.233579 -0.269999 -0.16903 0.0575098 0.178709 0.22156 0.426619 0.510001 0.696343 0.866632 0.961923 0.950291 0.975787 0.853998 0.799992 0.53724 0.359921 0.151191 0.247041 -0.0514688 0.0350768 -0.431708 -0.498329 -0.523113 -0.450042 -0.308131 -0.10889 0.0288939 -0.394679 -0.10217 -0.24005 0.0742572 0.27326 0.386471 0.590971 0.835862 0.742448 0.91842 0.823641 0.852431 0.9278 0.746554 0.704531 0.699291 0.531592 0.371369 0.195441 0.0550379 -0.01179 -0.15025 -0.23052 -0.254519 -0.443679 -0.611532 -0.688159 -0.713036 -0.823061 -0.676599 -0.585401 -0.426519 -0.573108 -0.25356
0 -0.63169 -0.367351 -0.51092 -0.752049 -0.849063 -0.72676 -0.48624 -0.1817 -0.0471209 -0.20698 -0.331411 -0.175389 -0.438409 -0.62124 -0.711451 -0.919457 -0.804968 -0.645023 -0.566912 -0.29766 0.0712382 0.19656 0.382179 0.10634 -0.050356 0.0651363 -0.311069 -0.303861 -0.53248 -0.735939 -0.873059 -0.843912 -0.936689 -0.933887 -0.827645 -0.670077 -0.451701 -0.351189 -0.0617087 0.2968 0.397029 0.604071 0.586972 0.324031 0.192931 0.26792 -0.0728137 -0.0945128 -0.551991 -0.69422 -0.830263 -0.925834 -0.964513 -0.983576 -0.992736 -0.881937 -0.773829 -0.630033 -0.446962 -0.230281 -0.0797927 0.160729 0.354309 0.36359 0.51847 0.489381 0.58283 0.779872 0.758972 0.509212 0.666138 0.39494 0.555451 0.42787 0.12048 0.0472653 -0.29788 -0.520971 -0.479052 -0.634258 -0.759272 -0.642379 -0.855722 -0.74349 -0.910222 -0.966977 -0.885898 -0.789172 -0.681409 -0.551187 -0.27374 -0.38801 -0.18608 0.01213 0.176319 0.17885

b-table from QSIPrep：

1 nan nan nan
999 0.618709 -0.317858 -0.718446
999 0.691545 -0.522141 -0.499134
999 0.526418 -0.533278 -0.662193
999 0.408959 -0.315281 -0.856359
999 0.451119 -0.092179 -0.887691
999 0.656195 -0.128340 -0.743597
999 0.781254 -0.314039 -0.539464
999 0.840153 -0.474177 -0.263248
999 0.712566 -0.674120 -0.194450
999 0.594251 -0.707067 -0.383304
999 0.440291 -0.723337 -0.531909
999 0.142891 -0.888078 -0.436921
999 0.257865 -0.710236 -0.655035
999 0.360226 -0.510700 -0.780656
999 0.132669 -0.462516 -0.876629
999 0.443319 0.167881 -0.880502
999 0.641997 0.141519 -0.753533
999 0.806156 0.155040 -0.571030
999 0.826923 -0.077380 -0.556966
999 0.915581 -0.245060 -0.318838
999 0.899319 -0.437280 0.003414
999 0.752104 -0.656482 0.058058
999 0.557734 -0.807060 0.193873
999 0.518685 -0.847531 -0.112501
999 0.343934 -0.888941 -0.302478
999 -0.030684 -0.972638 -0.230289
999 -0.362999 -0.733266 -0.574937
999 -0.073376 -0.815139 -0.574599
999 0.043666 -0.654223 -0.755040
999 -0.162445 -0.402057 -0.901089
999 -0.101452 -0.187179 -0.977073
999 0.189975 -0.250199 -0.949373
999 0.226272 -0.020620 -0.973846
999 0.367466 0.440778 -0.818952
999 0.567462 0.445903 -0.692212
999 0.738035 0.427102 -0.522386
999 0.906301 0.259461 -0.333613
999 0.952310 0.011620 -0.304910
999 0.979589 -0.191700 -0.060474
999 0.903251 -0.336519 0.266257
999 0.776124 -0.556336 0.296852
999 0.560207 -0.686634 0.463358
999 0.357750 -0.853438 0.379022
999 0.337324 -0.938538 0.073204
999 0.158187 -0.983230 -0.090753
999 -0.198077 -0.979623 -0.033222
999 -0.224305 -0.901280 -0.370650
999 -0.525623 -0.766203 -0.369666
999 -0.219320 -0.591074 -0.776228
999 -0.446227 -0.287897 -0.847347
999 -0.379670 -0.079921 -0.921663
999 -0.289892 0.142639 -0.946370
999 -0.036848 0.054140 -0.997853
999 -0.108488 0.347039 -0.931555
999 0.176972 0.257500 -0.949934
999 0.253687 0.664702 -0.702720
999 0.449749 0.688547 -0.568884
999 0.623845 0.662379 -0.414815
999 0.810525 0.529362 -0.250651
999 0.942850 0.324019 -0.077748
999 0.996203 0.086560 0.009276
999 0.973551 -0.116880 0.196312
999 0.832782 0.240840 0.498468
999 0.903168 -0.045660 0.426852
999 0.792006 0.042200 0.609053
999 0.834290 -0.253540 0.489568
999 0.719895 -0.467558 0.512972
999 0.432413 -0.623195 0.651649
999 0.209849 -0.803046 0.557745
999 0.151643 -0.958748 0.240429
999 0.009976 -0.911310 0.411600
999 -0.025182 -0.995019 0.096451
999 -0.170255 -0.947021 0.272331
999 -0.347492 -0.928247 0.132689
999 -0.382370 -0.904978 -0.186571
999 -0.683542 -0.696597 -0.217999
999 -0.641085 -0.551956 -0.533248
999 -0.472160 -0.484639 -0.736335
999 -0.679034 -0.329740 -0.655885
999 -0.644372 -0.130740 -0.753453
999 -0.574490 0.068480 -0.815642
999 -0.722078 0.173920 -0.669594
999 -0.460464 0.293780 -0.837655
999 -0.598369 0.397159 -0.695859
999 -0.279743 0.503438 -0.817493
999 0.082057 0.532564 -0.842403
999 -0.066920 0.691141 -0.719615
999 0.125764 0.811830 -0.570189
999 0.312757 0.847224 -0.429412
999 0.488272 0.824330 -0.286479
999 0.537192 0.843324 0.015164
999 0.668063 0.732566 -0.130533
999 0.815063 0.577825 0.042326
999 0.913056 0.362461 0.186954
999 0.802787 0.456403 0.383704
999 0.943716 0.142360 0.298553

Command

qsiprep /input /output participant --participant-label sub-KW92020230506 --output-resolution 2 --pepolar-method TOPUP --skip-bids-validation --anatomical-template MNI152NLin2009cAsym --work-dir /work --nprocs 16 --omp-nthreads 2

Errors

One or more execution steps failed (2). Error details are attached below.

Node Name: qsiprep_1_0_wf.sub_KW92020230506_wf.dwi_finalize_acq_dti96_dir_AP_wf.transform_dwis_t1.calculate_qc.raw_src
Node Name: qsiprep_1_0_wf.sub_KW92020230506_wf.dwi_finalize_acq_dti96_dir_AP_wf.final_denoise_wf.calculate_qc.raw_src
2

Hi @HanyuShao and welcome to neurostars!

What version are you using? And are you using a container (docker/singularity/apptainer)? What happens if you edit the raw b-vals so the first one is 0 and try rerunning in a new working directory? Does this error happen for all subjects in your dataset or just this one.

Best,
Steven

3

Dear Steven,

Thank you for your prompt and helpful suggestions. To follow up:

(1) I’m using the pennlinc/qsirecon Docker image (latest version).

(2) As you suspected, changing the first raw b-val to 0 resolved the issue, and the pipeline now runs successfully.

(3) This error appears consistently across all subjects with the same acquisition protocol (96 directions).

  Would you consider the modification (setting the first b-val to 0) a valid workaround, or is there a more robust way to address this? I’d greatly appreciate your insight.

   Thanks again for your support!
4

Hi @HanyuShao,

Great to hear that worked! I believe that the software is supposed to recognize any b-Val under 50 as a b0 volume by default, so this might be a bug. We can look into this next week.

Best,
Steven

5

@HanyuShao can you share the error logs from the failing nodes from your initial run?