Hi, We are having one scan going through recon-all that is taking way longer than ordinary. Other scans took about 20-30 hours to calculate, while this particular scan is taking 3+ days and has still not being completed. Just wondering how often something like this happens, and if there's any known problems with this sort of thing. Here's the mri_info output:
Starting DICOMRead2() dcmfile = /media/HIPPO/scans/normals/C11/T1/C11_0.dcm dcmdir = /media/HIPPO/scans/normals/C11/T1 Ref Series No = 3 Found 162 files, checking for dicoms Found 160 dicom files in series. First Sorting Computing Slice Direction Vs: 0 0 1 Vs: 0 0 1 Second Sorting Counting frames nframes = 1 nslices = 160 ndcmfiles = 160 PE Dir = ROW (dicom read) Volume information for C11_0.dcm type: dicom dimensions: 256 x 256 x 160 voxel sizes: 0.9375, 0.9375, 1.0000 type: SHORT (4) fov: 240.000 dof: 0 xstart: -128.0, xend: 128.0 ystart: -128.0, yend: 128.0 zstart: -80.0, zend: 80.0 TR: 11.96 msec, TE: 5.05 msec, TI: 300.00 msec, flip angle: 20.00 degrees nframes: 1 ras xform not present xform info: x_r = -1.0000, y_r = 0.0000, z_r = -0.0000, c_r = 5.3850 : x_a = 0.0000, y_a = -1.0000, z_a = -0.0000, c_a = 30.3850 : x_s = 0.0000, y_s = 0.0000, z_s = 1.0000, c_s = 27.4266 Orientation : ??? Primary Slice Direction: unknown
voxel to ras transform: -0.9375 0.0000 -0.0000 125.3850 0.0000 -0.9375 -0.0000 150.3850 0.0000 0.0000 1.0000 -52.5734 0.0000 0.0000 0.0000 1.0000
voxel-to-ras determinant 0.878906
ras to voxel transform: -1.0667 0.0000 0.0000 133.7440 0.0000 -1.0667 0.0000 160.4107 -0.0000 -0.0000 1.0000 52.5734 0.0000 0.0000 0.0000 1.0000
Hi David,
usually this means the topology fixing has encountered a very large defect, typically caused by the cerebellum or skull being attached to one of the cortical hemispheres. Try looking at the orig.nofix in tksurfer or the filled.mgz in tkmedit and do a sanity check on them.
cheers, Bruce On Tue, 15 Dec 2009, David Qixiang Chen wrote:
Hi, We are having one scan going through recon-all that is taking way longer than ordinary. Other scans took about 20-30 hours to calculate, while this particular scan is taking 3+ days and has still not being completed. Just wondering how often something like this happens, and if there's any known problems with this sort of thing. Here's the mri_info output:
Starting DICOMRead2() dcmfile = /media/HIPPO/scans/normals/C11/T1/C11_0.dcm dcmdir = /media/HIPPO/scans/normals/C11/T1 Ref Series No = 3 Found 162 files, checking for dicoms Found 160 dicom files in series. First Sorting Computing Slice Direction Vs: 0 0 1 Vs: 0 0 1 Second Sorting Counting frames nframes = 1 nslices = 160 ndcmfiles = 160 PE Dir = ROW (dicom read) Volume information for C11_0.dcm type: dicom dimensions: 256 x 256 x 160 voxel sizes: 0.9375, 0.9375, 1.0000 type: SHORT (4) fov: 240.000 dof: 0 xstart: -128.0, xend: 128.0 ystart: -128.0, yend: 128.0 zstart: -80.0, zend: 80.0 TR: 11.96 msec, TE: 5.05 msec, TI: 300.00 msec, flip angle: 20.00 degrees nframes: 1 ras xform not present xform info: x_r = -1.0000, y_r = 0.0000, z_r = -0.0000, c_r = 5.3850 : x_a = 0.0000, y_a = -1.0000, z_a = -0.0000, c_a = 30.3850 : x_s = 0.0000, y_s = 0.0000, z_s = 1.0000, c_s = 27.4266 Orientation : ??? Primary Slice Direction: unknown
voxel to ras transform: -0.9375 0.0000 -0.0000 125.3850 0.0000 -0.9375 -0.0000 150.3850 0.0000 0.0000 1.0000 -52.5734 0.0000 0.0000 0.0000 1.0000
voxel-to-ras determinant 0.878906
ras to voxel transform: -1.0667 0.0000 0.0000 133.7440 0.0000 -1.0667 0.0000 160.4107 -0.0000 -0.0000 1.0000 52.5734 0.0000 0.0000 0.0000 1.0000 _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Check the skull strip. Sometimes this happens when some structure is not removed. Also check whether the brain stem and cerebellum have been properly cut away and that the hemispheres have been properly separated (look at the wm.mgz for this).
David Qixiang Chen wrote:
Hi, We are having one scan going through recon-all that is taking way longer than ordinary. Other scans took about 20-30 hours to calculate, while this particular scan is taking 3+ days and has still not being completed. Just wondering how often something like this happens, and if there's any known problems with this sort of thing. Here's the mri_info output:
Starting DICOMRead2() dcmfile = /media/HIPPO/scans/normals/C11/T1/C11_0.dcm dcmdir = /media/HIPPO/scans/normals/C11/T1 Ref Series No = 3 Found 162 files, checking for dicoms Found 160 dicom files in series. First Sorting Computing Slice Direction Vs: 0 0 1 Vs: 0 0 1 Second Sorting Counting frames nframes = 1 nslices = 160 ndcmfiles = 160 PE Dir = ROW (dicom read) Volume information for C11_0.dcm type: dicom dimensions: 256 x 256 x 160 voxel sizes: 0.9375, 0.9375, 1.0000 type: SHORT (4) fov: 240.000 dof: 0 xstart: -128.0, xend: 128.0 ystart: -128.0, yend: 128.0 zstart: -80.0, zend: 80.0 TR: 11.96 msec, TE: 5.05 msec, TI: 300.00 msec, flip angle: 20.00 degrees nframes: 1 ras xform not present xform info: x_r = -1.0000, y_r = 0.0000, z_r = -0.0000, c_r = 5.3850 : x_a = 0.0000, y_a = -1.0000, z_a = -0.0000, c_a = 30.3850 : x_s = 0.0000, y_s = 0.0000, z_s = 1.0000, c_s = 27.4266 Orientation : ??? Primary Slice Direction: unknown
voxel to ras transform: -0.9375 0.0000 -0.0000 125.3850 0.0000 -0.9375 -0.0000 150.3850 0.0000 0.0000 1.0000 -52.5734 0.0000 0.0000 0.0000 1.0000
voxel-to-ras determinant 0.878906
ras to voxel transform: -1.0667 0.0000 0.0000 133.7440 0.0000 -1.0667 0.0000 160.4107 -0.0000 -0.0000 1.0000 52.5734 0.0000 0.0000 0.0000 1.0000 _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
where is it getting stuck? that is, look at the recon-all.log (and recon-all.cmd file). often times recon-all gets stuck in the topology fixer, because of some horrible wm.mgz file, which itself could be due to other problems. also, i assume the first thing performed before starting recon-all was to open the dicom in tkmedit (tkmedit -f name-of-first-slice.dcm) and check if it 'looks good' (that is, looks like all the other good images that you have processed).
n.
On Tue, 2009-12-15 at 17:15 -0500, David Qixiang Chen wrote:
Hi, We are having one scan going through recon-all that is taking way longer than ordinary. Other scans took about 20-30 hours to calculate, while this particular scan is taking 3+ days and has still not being completed. Just wondering how often something like this happens, and if there's any known problems with this sort of thing. Here's the mri_info output:
Starting DICOMRead2() dcmfile = /media/HIPPO/scans/normals/C11/T1/C11_0.dcm dcmdir = /media/HIPPO/scans/normals/C11/T1 Ref Series No = 3 Found 162 files, checking for dicoms Found 160 dicom files in series. First Sorting Computing Slice Direction Vs: 0 0 1 Vs: 0 0 1 Second Sorting Counting frames nframes = 1 nslices = 160 ndcmfiles = 160 PE Dir = ROW (dicom read) Volume information for C11_0.dcm type: dicom dimensions: 256 x 256 x 160 voxel sizes: 0.9375, 0.9375, 1.0000 type: SHORT (4) fov: 240.000 dof: 0 xstart: -128.0, xend: 128.0 ystart: -128.0, yend: 128.0 zstart: -80.0, zend: 80.0 TR: 11.96 msec, TE: 5.05 msec, TI: 300.00 msec, flip angle: 20.00 degrees nframes: 1 ras xform not present xform info: x_r = -1.0000, y_r = 0.0000, z_r = -0.0000, c_r = 5.3850 : x_a = 0.0000, y_a = -1.0000, z_a = -0.0000, c_a = 30.3850 : x_s = 0.0000, y_s = 0.0000, z_s = 1.0000, c_s = 27.4266 Orientation : ??? Primary Slice Direction: unknown
voxel to ras transform: -0.9375 0.0000 -0.0000 125.3850 0.0000 -0.9375 -0.0000 150.3850 0.0000 0.0000 1.0000 -52.5734 0.0000 0.0000 0.0000 1.0000
voxel-to-ras determinant 0.878906
ras to voxel transform: -1.0667 0.0000 0.0000 133.7440 0.0000 -1.0667 0.0000 160.4107 -0.0000 -0.0000 1.0000 52.5734 0.0000 0.0000 0.0000 1.0000 _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu
-
Bruce Fischl -
David Qixiang Chen -
Douglas N Greve -
Nick Schmansky