Hi Paige,

so I went and copied the following files from freesurferv4.4.0/bin to freesurferv3.0.5/bin: mri_convert mri_parse_sdcmdir mri_probedicom unpacksdcmdir tcl_setup unpacksdcmdir.tcl

I then edited the new unpacksdcmdir to read (you will have to put in the proper path to your freesurfer 4 install): #!/bin/tcsh -ef

set FREESURFER4_HOME = /space/opt/freesurfer-Linux-centos4_x86_64-stable-pub-v4.4.0-full/ source $FREESURFER_HOME/bin/tcl_setup.4.4.0 $FREESURFER4_HOME/lib/tcltktixblt/bin/tclsh8.4 $FREESURFER_HOME/bin/unpacksdcmdir.tcl.4.4.0 $argv

I hope that this will help you to get your old setup working with Numaris B17

best sebastian

On Mar 19, 2010, at 1:23 PM, Paige Scalf wrote:

Well, I have retinotopy data (analyzed using FSL) that is flatmapped and registered (painfully by hand) using 3.0.3. I need the data for my current study, and I'm not sure how well it would survive transformation to the more recent version. If someone can tell me that the data I processed using 3.0.3 will still be usable/play nicely after an upgrade, I can do that. If that's not possible, I'd prefer to transplant files; multiple operational versions of software does get confusing, but that would also work as a last resort.

Thanks, Paige

On Fri, Mar 19, 2010 at 3:10 PM, Sebastian Moeller sebastian.moeller@lur.rwth-aachen.de wrote: Hi Paige,

I ran into a similar problem a few months ago. Freesurfer 3.0.5 could not read data from a TIM trio with numaris B17. I solved this by transplanting (copying) a number of files from freesurfer 4.4.0 (the recent one at the time) into 3.0.5. unpacsdcmdir, mri_convert and a few more, that worked okay, but I since switched to freesurfer 4.5.0 for good. If you have to stick to 3.0.3 I can try to look up which files I had to replace in 3.0.5. Or alternatively you can install v4.5.0 parallel to 3.0.3 and only use it to import the DICOM data...

best regards Sebastian

On Mar 19, 2010, at 12:37 PM, freesurfer-request@nmr.mgh.harvard.edu wrote:

...

Message: 7 Date: Fri, 19 Mar 2010 14:37:24 -0500 From: Paige Scalf pscalf@gmail.com Subject: [Freesurfer] trio dicom files and mri_convert To: freesurfer@nmr.mgh.harvard.edu Message-ID: 1113e70c1003191237y6f5723aeha927124dd30b593e@mail.gmail.com Content-Type: text/plain; charset="iso-8859-1"

Hello,

I've been trying to use mri_convert (freesurfer-Darwin-tiger-stable-pub-v3.0.3) to reconstruct images from our Siemens Magnetom Trio. When I type:

mri_convert -it siemens_dicom -ot analyze4d DB_A0A1_318.MR.BECK-PAIGE_ A0A1_INHABITION.0003.0001.2010.03.18.16.34.50.500000.21615932.IMA Block1.img

DB*.IMA and Block1.img are the input and output files respectively, I get the output detailed below (the salient error seems to be that mri_convert can't find a useful transformation matrix).

Can you tell me what I can do to correct this problem? Many thanks in advance,

Paige Scalf

mri_convert -it siemens_dicom -ot analyze4d DB_A0A1_318.MR.BECK-PAIGE_A0A1_INHABITION.0003.0001.2010.03.18.16.34.50.500000.21615932.IMA Block1.img reading from DB_A0A1_318.MR.BECK-PAIGE_A0A1_INHABITION.0003.0001.2010.03.18.16.34.50.500000.21615932.IMA... Getting Series No Scanning Directory INFO: Found 111 files in /Volumes/dubois/A0A1inhibition/DB_A0A1_318/Block1 INFO: Scanning for Series Number 3 INFO: found 107 files in series INFO: loading series header info.

INFO: sorting. RunNo = 2 INFO: (120 110 50), nframes = 107, ismosaic=1 ERROR: incorreclty formatted version string syngo MR B17 found in dicom tag 18,1020 (len = 3 != 6) Numaris Version: syngo MR B17 Maj = 0, Min=-1073775504, MinMin = 0 mojo:/Volumes/dubois/A0A1inhibition/DB_A0A1_318/Block1 paige$ mri_convert -it dicom -ot analyze4d DB_A0A1_318.MR.BECK-PAIGE_A0A1_INHABITION.0003.0001.2010.03.18.16.34.50.500000.21615932.IMA Block1.img mri_convert -it dicom -ot analyze4d DB_A0A1_318.MR.BECK-PAIGE_A0A1_INHABITION.0003.0001.2010.03.18.16.34.50.500000.21615932.IMA Block1.img reading from DB_A0A1_318.MR.BECK-PAIGE_A0A1_INHABITION.0003.0001.2010.03.18.16.34.50.500000.21615932.IMA... Getting Series No Scanning Directory INFO: Found 111 files in /Volumes/dubois/A0A1inhibition/DB_A0A1_318/Block1 INFO: Scanning for Series Number 3 INFO: found 107 files in series INFO: loading series header info.

INFO: sorting. RunNo = 2 INFO: (120 110 50), nframes = 107, ismosaic=1 ERROR: incorreclty formatted version string syngo MR B17 found in dicom tag 18,1020 (len = 3 != 6) Numaris Version: syngo MR B17 Maj = -1073770984, Min=22021328, MinMin = -1610546680 107 DICOM 3.0 files in list Found 107 DICOM Files WARNING: NumberOfFrames 0 != Found Count of slices 107. reading DICOM image...

---

DICOM meta-header

file name /Volumes/dubois/A0A1inhibition/DB_A0A1_318/Block1/DB_A0A1_318.MR.BECK-PAIGE_A0A1_INHABITION.0003.0001.2010.03.18.16.34.50.500000.21615932.IMA Date and time study date 20100318 study time 160300.312000 series time 162802.453000 acquisition time 162802.453000 Identification patient name DB_A0A1_318 manufacturer SIEMENS Dimensions number of rows 880 number of columns 960 number of frames 107 pixel width 2.13333 pixel height 2.13333 slice thickness 2.1 field of view 1877.33 image number 1 (might be not reliable) transfer syntax UID 1.2.840.10008.1.2.1 Acquisition parameters echo time 25 repetition time 3000 inversion time not found echo number 1 flip angle 1.5708 bits allocated 16 Spatial information first image position -1027.46 -12.1727 952.833 last image position -1027.46 -12.1727 952.833 image orientation 1 0 0 0 -0.00174528 -0.999998

---

TR=3000.00, TE=25.00, TI=0.00, flip angle=90.00 i_ras = (-1, -0, 0) j_ras = (-0, 0.00174528, -0.999999) k_ras = (nan, nan, nan) Reslicing using trilinear interpolation MRIresample(): error inverting matrix; determinant is nan, matrix is: -2.133 -0.000 nan nan; -0.000 0.004 nan nan; 0.000 -2.133 nan nan; 0.000 0.000 0.000 1.000;