mne_organize_dicom not found when making flash bem

YuZhou · March 27, 2022, 7:57am

MNE version: MNE1.0.0
operating system: windows10

Hi all,
I’m trying to do the coregistration after the step of recon-all and make_scalp_surfaces(). Here are my code snippets of these two steps:

# recon-all
export SUBJECTS_DIR=/seastor/zhouyu0206/proj01/fmri/freesurfer/recon_all/
my_subject=subj-41
my_NIfTI=/brain/guixue/zhouyu/proj01/mri/nifti/sub-41/anat/sub-41_T1w.nii.gz
recon-all -i $my_NIfTI -s $my_subject -all

# make_scalp_surfaces
subject = 'subj-41'
subjects_dir = '/brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/'
mne.bem.make_scalp_surfaces(subject, subjects_dir=subjects_dir, force=True, overwrite=False, no_decimate=False, verbose=None)

and the result folder of reconstruction for this subject looked like this:

then I print ‘mne coreg’ in the terminal to initiate the coregistration GUI, and here is the error message I got:

Using pyvistaqt 3d backend.

Using high resolution head model in E:\proj01\analysis\fmri\freesurfer\recon_all\subj-41\bem\subj-41-head-dense.fif
Traceback (most recent call last):
  File "D:\anaconda3\envs\mne\Scripts\mne-script.py", line 9, in <module>
    sys.exit(main())
  File "D:\anaconda3\envs\mne\lib\site-packages\mne\commands\utils.py", line 107, in main
    cmd.run()
  File "D:\anaconda3\envs\mne\lib\site-packages\mne\commands\mne_coreg.py", line 85, in run
    mne.gui.coregistration(
  File "<decorator-gen-558>", line 12, in coregistration
  File "D:\anaconda3\envs\mne\lib\site-packages\mne\gui\__init__.py", line 193, in coregistration
    return CoregistrationUI(
  File "<decorator-gen-570>", line 12, in __init__
  File "D:\anaconda3\envs\mne\lib\site-packages\mne\gui\_coreg.py", line 238, in __init__
    self.coreg = Coregistration(
  File "D:\anaconda3\envs\mne\lib\site-packages\mne\coreg.py", line 1371, in __init__
    self._setup_bem()
  File "D:\anaconda3\envs\mne\lib\site-packages\mne\coreg.py", line 1426, in _setup_bem
    'mesh (%d vertices): %s' % (len(self._bem_high_res.surf.rr),
AttributeError: 'dict' object has no attribute 'surf'

then I thought maybe because I didn’t do the make_flash_bem step, this error message occurred.
so I start constructing the BEM surface. I had flash images for some of my subjects, here are the flash images files of one subject:

I followed the instruction in this page https://mne.tools/dev/generated/mne.bem.convert_flash_mris.html, which in Notes section says that

Before running this script do the following: (unless convert=False is specified)

Copy all of your FLASH images in a single directory and

create a directory to hold the output of mne_organize_dicom

cd to and run

$ mne_organize_dicom to create an appropriate directory structure

Create symbolic links to make flash05 and flash30 point to the

appropriate series: ln -s <FLASH 5 series dir> flash05 ln -s <FLASH 30 series dir> flash30 Some partition formats (e.g. FAT32) do not support symbolic links. In this case, copy the file to the appropriate series: cp <FLASH 5 series dir> flash05 cp <FLASH 30 series dir> flash30

cd to the directory where flash05 and flash30 links are

Set SUBJECTS_DIR and SUBJECT environment variables appropriately

Run this script

The fact is I stuck in the step 2,
mne_organize_dicom /brain/guixue/zhouyu/proj01/mri/nifti/sub-41/flash_source, the error I got is:
bash: mne_organize_dicom: command not found

I’m really confused about what should I do to solve this problem, any suggestions? Thanks a lot!

Best

agramfort · March 27, 2022, 5:09pm

did you modify manually some files in the recon-all output subject folder?

Alex

YuZhou · March 28, 2022, 1:35am

Hi! Alex,
Thank you for your reply
I didn’t manually modify any files in the recon-all output subject folder. Taking sub-41 as an example, after the recon-all, the output subject folder structure looked like this except the ‘bem’ folder which didn’t exist until the make_scalp_surfaces() step.

and my nifti folder looked like this:

in this case, the flash dicoms were already converted to .nii.gz format.
1)Should I convert the flash DICOM to .mgz format?
2) I followed the instructions of the Notes section mne.bem.convert_flash_mris — MNE 1.1.dev0 documentation despite the step 2 didn’t work, I create the flash05 and flash30 symbolic links in the folder: ‘flash_dest’

and then run mne flash_bem --subject=sub-41 dir=/brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all --view,
the error prompted that

---- Converting Flash images ----
Traceback (most recent call last):
  File "/brain/guixue/.conda/envs/mne/bin/mne", line 8, in <module>
    sys.exit(main())
  File "/brain/guixue/.conda/envs/mne/lib/python3.8/site-packages/mne/commands/utils.py", line 107, in main
    cmd.run()
  File "/brain/guixue/.conda/envs/mne/lib/python3.8/site-packages/mne/commands/mne_flash_bem.py", line 100, in run
    convert_flash_mris(subject=subject, subjects_dir=subjects_dir,
  File "<decorator-gen-69>", line 22, in convert_flash_mris
  File "/brain/guixue/.conda/envs/mne/lib/python3.8/site-packages/mne/bem.py", line 1759, in convert_flash_mris
    raise RuntimeError("Directory %s is missing."
RuntimeError: Directory flash05/002 is missing.

I assumed this error resulted from the failure in step 2 where the
mne_organize_dicom: command not found, so the folder structure was wrong.
It’s kind of a mess in my mind now, what should I do

agramfort · March 28, 2022, 8:32am

if you have .nii files you don’t need the mne_organize_dicom command

I am a bit lost here

Alex

richard · March 28, 2022, 10:38am

+1 on this. You should only need to run recon-all, and this should produce everything you need to get to work with MNE-Python… (well, except for the BEM and head surface, but you already know how to create those)

YuZhou · March 28, 2022, 11:06am

My biggest confusion is about how to create the flash bem. In my subject folders of recon-all output, there seem no flash image files in $SUBJECTS_DIR/$SUBJECT/mri/flash/parameter_maps, which means that the recon-all didn’t process the flash .nii.gz files in my nifiti folder. Or are the flash .nii.gz files in my nifiti folder is all the mne flash_bem needs?

richard · March 28, 2022, 12:05pm

That’s the problem then. I’m not sure how you ended up this way, but the directory must not be empty if you want to use FLASH images…

YuZhou · March 28, 2022, 12:13pm

Here is the original DICOM folder structure of one subject (sub-41) in which there were 4 folders of flash images:

and this is the folder structure of nfiti files converted from DICOMs:

since recon-all didn’t process the flash images, would it be there is something wrong with these flash images?

agramfort · March 28, 2022, 1:29pm

what you looked the doc of https://mne.tools/stable/generated/mne.bem.make_flash_bem.html

flash_path parameter may help?

recon-all does not do anything with the flash images

ALex

YuZhou · March 28, 2022, 1:54pm

Hi! Alex,
I tried your suggestion, glad we move forward a little bit because another error occurred, here is my code:

subject = 'sub-41'

subjects_dir = '/seastor/zhouyu0206/proj01/fmri/freesurfer/recon_all/'

flash_path = '/seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/'

mne.bem.make_flash_bem(subject, overwrite=False, show=True, subjects_dir=subjects_dir, flash_path=flash_path, copy=False, verbose=None)

here is the error message (sorry it’s long ):

Processing the flash MRI data to produce BEM meshes with the following parameters:
SUBJECTS_DIR = /seastor/zhouyu0206/proj01/fmri/freesurfer/recon_all
SUBJECT = sub-41
Result dir = /seastor/zhouyu0206/proj01/fmri/freesurfer/recon_all/sub-41/bem/flash


---- Registering flash 5 with MPRAGE ----
Running subprocess: fsl_rigid_register -r /seastor/zhouyu0206/proj01/fmri/freesurfer/recon_all/sub-41/mri/T1.mgz -i /seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/flash5.mgz -o /seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/flash5_reg.mgz






$Id: fsl_rigid_register,v 1.33 2016/02/16 17:17:20 zkaufman Exp $

Mon 28 Mar 2022 09:49:00 PM CST

/seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/flash5_reg.mgz

--------------------------------------

/tmp/tmp_mne_tempdir_bldfvslo

mri_convert /seastor/zhouyu0206/proj01/fmri/freesurfer/recon_all/sub-41/mri/T1.mgz /seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/fsl_rigid_register.22494/refvol.fsl_rigid_register.nii

mri_convert.bin /seastor/zhouyu0206/proj01/fmri/freesurfer/recon_all/sub-41/mri/T1.mgz /seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/fsl_rigid_register.22494/refvol.fsl_rigid_register.nii 

$Id: mri_convert.c,v 1.226 2016/02/26 16:15:24 mreuter Exp $

reading from /seastor/zhouyu0206/proj01/fmri/freesurfer/recon_all/sub-41/mri/T1.mgz...

TR=2530.00, TE=0.00, TI=0.00, flip angle=0.00

i_ras = (-1, 9.84528e-11, 0)

j_ras = (0, 0, -1)

k_ras = (1.34378e-10, 1, 3.63798e-12)

writing to /seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/fsl_rigid_register.22494/refvol.fsl_rigid_register.nii...

--------------------------------------

/tmp/tmp_mne_tempdir_bldfvslo

mri_convert /seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/flash5.mgz /seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/fsl_rigid_register.22494/invol.fsl_rigid_register.nii

mri_convert.bin /seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/flash5.mgz /seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/fsl_rigid_register.22494/invol.fsl_rigid_register.nii 

$Id: mri_convert.c,v 1.226 2016/02/26 16:15:24 mreuter Exp $

reading from /seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/flash5.mgz...

mghRead(/seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/flash5.mgz, -1): could not open file

('\n\n\n$Id: fsl_rigid_register,v 1.33 2016/02/16 17:17:20 zkaufman Exp $\nMon 28 Mar 2022 09:49:00 PM CST\n/seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/flash5_reg.mgz\n--------------------------------------\n/tmp/tmp_mne_tempdir_bldfvslo\nmri_convert /seastor/zhouyu0206/proj01/fmri/freesurfer/recon_all/sub-41/mri/T1.mgz /seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/fsl_rigid_register.22494/refvol.fsl_rigid_register.nii\nmri_convert.bin /seastor/zhouyu0206/proj01/fmri/freesurfer/recon_all/sub-41/mri/T1.mgz /seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/fsl_rigid_register.22494/refvol.fsl_rigid_register.nii \n$Id: mri_convert.c,v 1.226 2016/02/26 16:15:24 mreuter Exp $\nreading from /seastor/zhouyu0206/proj01/fmri/freesurfer/recon_all/sub-41/mri/T1.mgz...\nTR=2530.00, TE=0.00, TI=0.00, flip angle=0.00\ni_ras = (-1, 9.84528e-11, 0)\nj_ras = (0, 0, -1)\nk_ras = (1.34378e-10, 1, 3.63798e-12)\nwriting to /seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/fsl_rigid_register.22494/refvol.fsl_rigid_register.nii...\n--------------------------------------\n/tmp/tmp_mne_tempdir_bldfvslo\nmri_convert /seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/flash5.mgz /seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/fsl_rigid_register.22494/invol.fsl_rigid_register.nii\nmri_convert.bin /seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/flash5.mgz /seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/fsl_rigid_register.22494/invol.fsl_rigid_register.nii \n$Id: mri_convert.c,v 1.226 2016/02/26 16:15:24 mreuter Exp $\nreading from /seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/flash5.mgz...\n', 'mghRead(/seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/flash5.mgz, -1): could not open file\n')
---------------------------------------------------------------------------
CalledProcessError                        Traceback (most recent call last)
Input In [4], in <cell line: 1>()
----> 1 mne.bem.make_flash_bem(subject, overwrite=False, show=True, subjects_dir=subjects_dir, flash_path=flash_path, copy=False, verbose=None)

File <decorator-gen-76>:12, in make_flash_bem(subject, overwrite, show, subjects_dir, flash_path, copy, verbose)

File ~/anaconda2/envs/mne/lib/python3.10/site-packages/mne/bem.py:1906, in make_flash_bem(subject, overwrite, show, subjects_dir, flash_path, copy, verbose)
   1903         ref_volume = op.join(mri_dir, 'T1')
   1904     cmd = ['fsl_rigid_register', '-r', ref_volume, '-i', flash5,
   1905            '-o', flash5_reg]
-> 1906     run_subprocess_env(cmd)
   1907 else:
   1908     logger.info("Registered flash 5 image is already there")

File <decorator-gen-1>:12, in run_subprocess(command, return_code, verbose, *args, **kwargs)

File ~/anaconda2/envs/mne/lib/python3.10/site-packages/mne/utils/misc.py:170, in run_subprocess(command, return_code, verbose, *args, **kwargs)
    168 err_fun = subprocess.CalledProcessError.__init__
    169 if 'output' in _get_args(err_fun):
--> 170     raise subprocess.CalledProcessError(p.returncode, command, output)
    171 else:
    172     raise subprocess.CalledProcessError(p.returncode, command)

CalledProcessError: Command '['fsl_rigid_register', '-r', '/seastor/zhouyu0206/proj01/fmri/freesurfer/recon_all/sub-41/mri/T1.mgz', '-i', '/seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/flash5.mgz', '-o', '/seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/flash5_reg.mgz']' returned non-zero exit status 1.

my flash images folder looks like this:

my subject recon-all output folder looks like this:

agramfort · March 29, 2022, 9:14am

the error says:

/seastor/zhouyu0206/proj01/fmri/nifti/sub-41/flash/flash5.mgz

is not found.

So the flash5 file should be called like this. I think the API
is really rough on the edges but I least it should help

Alex

YuZhou · March 29, 2022, 11:51am

Hi! Alex and Richard,
I think I kind of found a solution to process my flash images. The reason for the recon-all didn’t process my flash images might be they are multi-echo Flash sequence (MEF) data while FreeSurfer is not optimized for the processing of MEF data yet, but it did give two solutions here MEF - Free Surfer Wiki
and I used the second one from step1-3. and then use the mne.bem.make_flash_bem(subject, overwrite=False, show=True, subjects_dir=subjects_dir, flash_path=None, copy=False, verbose=None), it created the inner skull, outer skull and outer skin surface file, but something went wrong when plot the bem, here is output folder structure and the verbose:

Processing the flash MRI data to produce BEM meshes with the following parameters:
SUBJECTS_DIR = /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all
SUBJECT = sub-41-1
Result dir = /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/bem/flash


---- Registering flash 5 with MPRAGE ----
Registered flash 5 image is already there

---- Converting flash5 volume into COR format ----
Running subprocess: mri_convert /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/flash/parameter_maps/flash5_reg.mgz /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/flash5
mri_convert.bin /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/flash/parameter_maps/flash5_reg.mgz /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/flash5 

$Id: mri_convert.c,v 1.226 2016/02/26 16:15:24 mreuter Exp $

reading from /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/flash/parameter_maps/flash5_reg.mgz...

TR=2530.00, TE=0.00, TI=0.00, flip angle=0.00

i_ras = (-1, 9.84528e-11, 0)

j_ras = (0, 0, -1)

k_ras = (1.34378e-10, 1, 3.63798e-12)

changing data type from float to uchar (noscale = 0)...

MRIchangeType: Building histogram 

writing to /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/flash5...


---- Converting T1 volume into COR format ----
Running subprocess: mri_convert /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/T1.mgz /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/T1
mri_convert.bin /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/T1.mgz /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/T1 

$Id: mri_convert.c,v 1.226 2016/02/26 16:15:24 mreuter Exp $

reading from /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/T1.mgz...

TR=2530.00, TE=0.00, TI=0.00, flip angle=0.00

i_ras = (-1, 9.84528e-11, 0)

j_ras = (0, 0, -1)

k_ras = (1.34378e-10, 1, 3.63798e-12)

writing to /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/T1...


---- Converting brain volume into COR format ----
Running subprocess: mri_convert /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/brain.mgz /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/brain
mri_convert.bin /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/brain.mgz /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/brain 

$Id: mri_convert.c,v 1.226 2016/02/26 16:15:24 mreuter Exp $

reading from /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/brain.mgz...

TR=2530.00, TE=0.00, TI=0.00, flip angle=0.00

i_ras = (-1, 9.84528e-11, 0)

j_ras = (0, 0, -1)

k_ras = (1.34378e-10, 1, 3.63798e-12)

writing to /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/brain...


---- Creating the BEM surfaces ----
Running subprocess: mri_make_bem_surfaces sub-41-1
nvertices=2562, nfaces=5120

mri_strip_skull: triangle file /opt/software/freesurfer-Linux-centos6_x86_64-stable-pub-v6.0.0/freesurfer/lib/bem/ic4.tri read

mri_strip_skull: images /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/brain/COR- read

xlo=-71.000000, xhi=70.000000, ylo=-84.000000, yhi=76.000000, zlo=-34.000000, zhi=103.000000

0: sval=84.06,sinval=84.06,soutval=84.06

10: sval=69.20,sinval=69.20,soutval=69.20

20: sval=30.72,sinval=30.72,soutval=30.72

30: sval=14.96,sinval=14.96,soutval=14.96

40: sval=13.41,sinval=13.41,soutval=13.41

50: sval=13.98,sinval=13.98,soutval=13.98

60: sval=14.17,sinval=14.17,soutval=14.17

70: sval=13.52,sinval=13.52,soutval=13.52

80: sval=13.51,sinval=13.51,soutval=13.51

90: sval=13.45,sinval=13.45,soutval=13.45

0: ad=0.520972, dmax=1.000000, nclip=0

1: ad=0.478029, dmax=1.000000, nclip=0

2: ad=0.488134, dmax=1.000000, nclip=0

3: ad=0.501572, dmax=1.000000, nclip=0

4: ad=0.490254, dmax=1.000000, nclip=0

5: ad=0.473747, dmax=1.000000, nclip=0

6: ad=0.483116, dmax=1.000000, nclip=0

7: ad=0.491565, dmax=1.000000, nclip=0

8: ad=0.488759, dmax=1.000000, nclip=0

9: ad=0.480460, dmax=1.000000, nclip=0

10: ad=0.497451, dmax=1.000000, nclip=0

11: ad=0.512828, dmax=1.000000, nclip=0

12: ad=0.508317, dmax=1.000000, nclip=0

13: ad=0.502193, dmax=1.000000, nclip=0

14: ad=0.526212, dmax=1.000000, nclip=0

15: ad=0.540898, dmax=1.000000, nclip=0

16: ad=0.524475, dmax=1.000000, nclip=0

17: ad=0.531350, dmax=1.000000, nclip=0

18: ad=0.563880, dmax=1.000000, nclip=0

19: ad=0.560443, dmax=1.000000, nclip=0

20: ad=0.543994, dmax=1.000000, nclip=0

21: ad=0.564087, dmax=1.000000, nclip=0

22: ad=0.594574, dmax=1.000000, nclip=0

23: ad=0.576289, dmax=1.000000, nclip=0

24: ad=0.564603, dmax=1.000000, nclip=0

25: ad=0.596765, dmax=1.000000, nclip=0

26: ad=0.614337, dmax=1.000000, nclip=0

27: ad=0.588984, dmax=1.000000, nclip=0

28: ad=0.588496, dmax=1.000000, nclip=0

29: ad=0.619174, dmax=1.000000, nclip=0

30: ad=0.627111, dmax=1.000000, nclip=0

31: ad=0.597021, dmax=1.000000, nclip=0

32: ad=0.610698, dmax=1.000000, nclip=0

33: ad=0.634660, dmax=1.000000, nclip=0

34: ad=0.632556, dmax=1.000000, nclip=0

35: ad=0.611053, dmax=1.000000, nclip=0

36: ad=0.631502, dmax=1.000000, nclip=0

37: ad=0.644499, dmax=1.000000, nclip=0

38: ad=0.634269, dmax=1.000000, nclip=0

39: ad=0.626527, dmax=1.000000, nclip=0

40: ad=0.644340, dmax=1.000000, nclip=0

41: ad=0.648393, dmax=1.000000, nclip=0

42: ad=0.634235, dmax=1.000000, nclip=0

43: ad=0.640851, dmax=1.000000, nclip=0

44: ad=0.653961, dmax=1.000000, nclip=0

45: ad=0.650836, dmax=1.000000, nclip=0

46: ad=0.636601, dmax=1.000000, nclip=0

47: ad=0.652522, dmax=1.000000, nclip=0

48: ad=0.660308, dmax=1.000000, nclip=0

49: ad=0.651074, dmax=1.000000, nclip=0

0: sval=29.99,sinval=29.99,soutval=29.99

10: sval=14.10,sinval=14.10,soutval=14.10

20: sval=13.84,sinval=13.84,soutval=13.84

30: sval=13.31,sinval=13.31,soutval=13.31

40: sval=13.14,sinval=13.14,soutval=13.14

50: sval=13.90,sinval=13.90,soutval=13.90

60: sval=13.67,sinval=13.67,soutval=13.67

70: sval=13.92,sinval=13.92,soutval=13.92

80: sval=13.24,sinval=13.24,soutval=13.24

90: sval=13.85,sinval=13.85,soutval=13.85

0: ad=0.407971, dmax=1.000000, nclip=0

1: ad=0.228519, dmax=1.000000, nclip=0

2: ad=0.100772, dmax=0.958336, nclip=0

3: ad=0.077238, dmax=0.762616, nclip=0

4: ad=0.054178, dmax=0.687747, nclip=0

5: ad=0.046000, dmax=0.597683, nclip=0

6: ad=0.039925, dmax=0.528905, nclip=0

7: ad=0.035881, dmax=0.477430, nclip=0

8: ad=0.032955, dmax=0.434377, nclip=0

9: ad=0.030635, dmax=0.396509, nclip=0

mri_strip_skull: triangle file /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/bem/inner_skull_tmp.tri written

nvertices=2562, nfaces=5120

mri_strip_skull: triangle file /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/bem/inner_skull_tmp.tri read

mri_strip_skull: images /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/flash5/COR- read

0: sval=73.94,sinval=73.94,soutval=73.94

10: sval= 0.00,sinval= 0.00,soutval= 0.00

20: sval= 0.00,sinval= 0.00,soutval= 0.00

30: sval= 0.00,sinval= 0.00,soutval= 0.00

mri_strip_skull: triangle file /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/bem/inner_skull.tri written

mri_strip_skull: triangle file /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/bem/outer_skull.tri written

nvertices=2562, nfaces=5120

mri_strip_skull: triangle file /opt/software/freesurfer-Linux-centos6_x86_64-stable-pub-v6.0.0/freesurfer/lib/bem/ic4.tri read

mri_strip_skull: images /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/mri/T1/COR- read

xlo=-82.000000, xhi=79.000000, ylo=-94.000000, yhi=102.000000, zlo=-102.000000, zhi=103.000000

0: sval= 4.89,sinval= 4.89,soutval= 4.89

10: sval= 5.85,sinval= 5.85,soutval= 5.85

20: sval= 7.32,sinval= 7.32,soutval= 7.32

30: sval=10.84,sinval=10.84,soutval=10.84

40: sval=28.12,sinval=28.12,soutval=28.12

50: sval=45.18,sinval=45.18,soutval=45.18

60: sval=58.08,sinval=58.08,soutval=58.08

70: sval=65.99,sinval=65.99,soutval=65.99

80: sval=70.12,sinval=70.12,soutval=70.12

90: sval=71.47,sinval=71.47,soutval=71.47

100: sval=70.91,sinval=70.91,soutval=70.91

110: sval=71.80,sinval=71.80,soutval=71.80

120: sval=71.57,sinval=71.57,soutval=71.57

130: sval=71.73,sinval=71.73,soutval=71.73

140: sval=71.31,sinval=71.31,soutval=71.31

150: sval=71.03,sinval=71.03,soutval=71.03

160: sval=70.77,sinval=70.77,soutval=70.77

170: sval=70.74,sinval=70.74,soutval=70.74

180: sval=70.79,sinval=70.79,soutval=70.79

190: sval=70.80,sinval=70.80,soutval=70.80

0: sval=71.49,sinval=71.49,soutval=71.49

10: sval=66.57,sinval=66.57,soutval=66.57

20: sval=66.01,sinval=66.01,soutval=66.01

30: sval=65.63,sinval=65.63,soutval=65.63

40: sval=66.25,sinval=66.25,soutval=66.25

50: sval=66.34,sinval=66.34,soutval=66.34

60: sval=65.21,sinval=65.21,soutval=65.21

70: sval=65.40,sinval=65.40,soutval=65.40

80: sval=65.13,sinval=65.13,soutval=65.13

90: sval=65.33,sinval=65.33,soutval=65.33

mri_strip_skull: triangle file /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/bem/outer_skin.tri written


---- Converting the tri files into surf files ----
Loaded surface from /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/bem/flash/inner_skull.tri with 2562 nodes and 5120 triangles.
Node normals were not included in the source file.
Loaded surface from /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/bem/flash/outer_skull.tri with 2562 nodes and 5120 triangles.
Node normals were not included in the source file.
Loaded surface from /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/bem/flash/outer_skin.tri with 2562 nodes and 5120 triangles.
Node normals were not included in the source file.

---- Cleaning up ----
Deleted the T1 COR volume
Deleted the brain COR volume
Deleted the flash5 COR volume

---- Creating symbolic links ----
Symbolic links to .surf files created in bem folder

Thank you for waiting.
The BEM triangulations for this subject are now available at:
/brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/bem.
We hope the BEM meshes created will facilitate your MEG and EEG data analyses.
Using surface: /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/bem/inner_skull.surf
Using surface: /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/bem/outer_skull.surf
Using surface: /brain/guixue/zhouyu/proj01/mri/freesurfer/recon_all/sub-41-1/bem/outer_skin.surf
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
Input In [21], in <cell line: 1>()
----> 1 mne.bem.make_flash_bem(subject, overwrite=False, show=True, subjects_dir=subjects_dir, flash_path=None, copy=False, verbose=None)

File <decorator-gen-76>:12, in make_flash_bem(subject, overwrite, show, subjects_dir, flash_path, copy, verbose)

File ~/.conda/envs/mne/lib/python3.9/site-packages/mne/bem.py:2003, in make_flash_bem(subject, overwrite, show, subjects_dir, flash_path, copy, verbose)
   2001 # Show computed BEM surfaces
   2002 if show:
-> 2003     plot_bem(subject=subject, subjects_dir=subjects_dir,
   2004              orientation='coronal', slices=None, show=True)

File ~/.conda/envs/mne/lib/python3.9/site-packages/mne/viz/misc.py:600, in plot_bem(subject, subjects_dir, orientation, slices, brain_surfaces, src, show, show_indices, mri, show_orientation)
    596     raise IOError('No surface files found. Surface files must end with '
    597                   'inner_skull.surf, outer_skull.surf or outer_skin.surf')
    599 # Plot the contours
--> 600 fig = _plot_mri_contours(
    601     mri_fname=mri_fname, surfaces=surfaces, src=src,
    602     orientation=orientation, slices=slices, show=show,
    603     show_indices=show_indices, show_orientation=show_orientation,
    604     slices_as_subplots=True
    605 )
    606 return fig

File ~/.conda/envs/mne/lib/python3.9/site-packages/mne/viz/misc.py:434, in _plot_mri_contours(mri_fname, surfaces, src, orientation, slices, show, show_indices, show_orientation, width, slices_as_subplots)
    432     with warnings.catch_warnings(record=True):  # ignore contour warn
    433         warnings.simplefilter('ignore')
--> 434         ax.tricontour(surf['rr'][:, x], surf['rr'][:, y],
    435                       surf['tris'], surf['rr'][:, axis],
    436                       levels=[sl], colors=color, linewidths=1.0,
    437                       zorder=1)
    439 if len(sources):
    440     in_slice = (sources[:, axis] >= lower) & (sources[:, axis] < upper)

File ~/.conda/envs/mne/lib/python3.9/site-packages/matplotlib/tri/tricontour.py:281, in tricontour(ax, *args, **kwargs)
    258 """
    259 %(_tricontour_doc)s
    260 
   (...)
    278     number of contour levels it will be repeated as necessary.
    279 """
    280 kwargs['filled'] = False
--> 281 return TriContourSet(ax, *args, **kwargs)

File ~/.conda/envs/mne/lib/python3.9/site-packages/matplotlib/tri/tricontour.py:29, in TriContourSet.__init__(self, ax, *args, **kwargs)
     19 def __init__(self, ax, *args, **kwargs):
     20     """
     21     Draw triangular grid contour lines or filled regions,
     22     depending on whether keyword arg 'filled' is False
   (...)
     27     are described in the docstring of `~.Axes.tricontour`.
     28     """
---> 29     super().__init__(ax, *args, **kwargs)

File ~/.conda/envs/mne/lib/python3.9/site-packages/matplotlib/contour.py:812, in ContourSet.__init__(self, ax, levels, filled, linewidths, linestyles, hatches, alpha, origin, extent, cmap, colors, norm, vmin, vmax, extend, antialiased, nchunk, locator, transform, *args, **kwargs)
    808     self.origin = mpl.rcParams['image.origin']
    810 self._transform = transform
--> 812 kwargs = self._process_args(*args, **kwargs)
    813 self._process_levels()
    815 self._extend_min = self.extend in ['min', 'both']

File ~/.conda/envs/mne/lib/python3.9/site-packages/matplotlib/tri/tricontour.py:45, in TriContourSet._process_args(self, *args, **kwargs)
     43 else:
     44     from matplotlib import _tri
---> 45     tri, z = self._contour_args(args, kwargs)
     46     C = _tri.TriContourGenerator(tri.get_cpp_triangulation(), z)
     47     self._mins = [tri.x.min(), tri.y.min()]

File ~/.conda/envs/mne/lib/python3.9/site-packages/matplotlib/tri/tricontour.py:72, in TriContourSet._contour_args(self, args, kwargs)
     69     raise ValueError('z must not contain masked points within the '
     70                      'triangulation')
     71 if not np.isfinite(z_check).all():
---> 72     raise ValueError('z array must not contain non-finite values '
     73                      'within the triangulation')
     75 z = np.ma.masked_invalid(z, copy=False)
     76 self.zmax = float(z_check.max())

ValueError: z array must not contain non-finite values within the triangulation

agramfort · March 29, 2022, 8:34pm

this image is not from a flash bem. The outer skull is still close to the skin.

Alex

YuZhou · March 30, 2022, 1:36am

Yeah, you’re right. When I tried to compare the bem plot from watershed and flash, I found them exactly same. It turned out that the plot_bem used the watershed results which together with the flash results in the same subject bem folder. When I remove the watershed and flash bem results from the folder, and run mne.bem.make_flash_bem(subject, overwrite=False, show=True, subjects_dir=subjects_dir, flash_path=None, copy=False, verbose=None), the error message showed up which I posted above.

agramfort · March 30, 2022, 7:59am

is there a way to access the data?

Alex

YuZhou · March 30, 2022, 10:04am

I tried Dropbox link, hope this work

these are DICOM files of the flash sequence

agramfort · March 31, 2022, 12:44pm

are you sure you have flash5 images? your image looks very different from the flash5 in
the MNE sample data

Alex

YuZhou · March 31, 2022, 1:05pm

Could it be the way of processing the flash dicoms wrong or the flash sequence scanner parameters wrong?
Because when I want to convert the flash dicoms following the MNE tutorials, like I said above the mne_organize_dicom: command not found and the future steps weren’t be continued so I tried other ways to process the flash images I posted above, I’m not sure it’s the problem of the processing of the flash images or the original scanner parameters might be wrong?

agramfort · April 1, 2022, 6:41am

can you try to convert the DICOM to a .nii or .mgz file using freesurfer?

look in the MNE sample dataset to see how the flash image should look like
in terms of contrast

Alex

YuZhou · April 1, 2022, 7:02am

Sure, I can try that In another hand, I’m also considering using Fieldtrip to create the BEMs because I read about better reconstructions from Fieldtrip using T1 scans and not all of the subjects have the flash images, but I haven’t dug into it. Will it be trouble to use the BEM from Fieldtrip in MNE?

Topic		Replies	Views
Problem with mne.gui.coregistration Mailing List Archive (read-only) list-archive	0	235	June 14, 2019
No standard head model found Support & Discussions coregistration	1	327	May 10, 2022
Issue with .surf files for mne.giu.coregistration - endings are missing + Attribute Error for self._bem_high_res.surf.rr Support & Discussions source-localization , coregistration , freesurfer , bem	1	307	March 17, 2022
Coregistration of my electrode locations with individual MRI images Support & Discussions eeg , coregistration	6	708	May 13, 2022
Issues with make_scalp_surfaces Support & Discussions coregistration	25	1859	May 11, 2022

mne_organize_dicom not found when making flash bem

Related topics