Dear Freesurfer/Tracula users, Finally I could successfully test Tracula and I think it is amazing what it can do automatically. These are the commands I used:
export FREESURFER_HOME=${HOME}/Programs/freesurfer_510 && source $FREESURFER_HOME/SetUpFreeSurfer.sh source /etc/fsl/fsl.sh export FSLPARALLEL=condor trac-all -prep -c /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/dmrirc_Fernando_Ventrice trac-all -bedp -c /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/dmrirc_Fernando_Ventrice # "trac-all -bedp" gave some errors so I used bedpostx directly: bedpostx /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/20120112_VENTRICE_FERNANDO/dmri trac-all -path -c /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/dmrirc_Fernando_Ventrice freeview -tv /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/20120112_VENTRICE_FERNANDO/dpath/merged_avg33_mni_flt.mgz -v /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/20120112_VENTRICE_FERNANDO/dmri/dtifit_FA.nii.gz &
All finished without errors, but there seems to be a problem with CC Forceps Mayor volume, which gave very small values. Is this OK or I made some mistake? First I tested with some patient and other fiber tracks also gave very small volumes. I attached some pictures and the dmrirc and trac-all log files. Best, Fernando Ventrice
El contenido del presente mensaje y el de sus adjuntos, es confidencial, privado y de uso exclusivo de los destinatarios a los cuales está dirigido, pudiendo contener información legalmente protegida. Queda prohibida la revisión, divulgación, publicación, modificación, copia, distribución o acción en relación con esta información, por personas o entidades distintas al destinatario. Si usted, por alguna razón, lo ha recibido por error, por favor tenga la amabilidad de eliminarlo de su sistema. Las opiniones contenidas, son exclusivas de su autor, y no representa necesariamente, la opinión de FLENI. http://www.fleni.org.ar
There were recently a few messages on the list regarding this. What you should do is include "set reinit=1" in your config file, and then run trac-all -priors -c $CONFIG trac-all -path -c $CONFIG
Make sure you only include the pathway of interest in the config file; otherwise, it will re-run the analysis for all 18. That has worked for me most of the time. ________________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edu [freesurfer-bounces@nmr.mgh.harvard.edu] on behalf of Fernando Ventrice [fventrice@fleni.org.ar] Sent: Thursday, October 18, 2012 10:14 AM To: freesurfer@nmr.mgh.harvard.edu Subject: [Freesurfer] FW: TRACULA some fiber volumes being too small
Dear Freesurfer/Tracula users, Finally I could successfully test Tracula and I think it is amazing what it can do automatically. These are the commands I used:
export FREESURFER_HOME=${HOME}/Programs/freesurfer_510 && source $FREESURFER_HOME/SetUpFreeSurfer.sh source /etc/fsl/fsl.sh export FSLPARALLEL=condor trac-all -prep -c /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/dmrirc_Fernando_Ventrice trac-all -bedp -c /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/dmrirc_Fernando_Ventrice # "trac-all -bedp" gave some errors so I used bedpostx directly: bedpostx /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/20120112_VENTRICE_FERNANDO/dmri trac-all -path -c /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/dmrirc_Fernando_Ventrice freeview -tv /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/20120112_VENTRICE_FERNANDO/dpath/merged_avg33_mni_flt.mgz -v /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/20120112_VENTRICE_FERNANDO/dmri/dtifit_FA.nii.gz &
All finished without errors, but there seems to be a problem with CC Forceps Mayor volume, which gave very small values. Is this OK or I made some mistake? First I tested with some patient and other fiber tracks also gave very small volumes. I attached some pictures and the dmrirc and trac-all log files. Best, Fernando Ventrice
El contenido del presente mensaje y el de sus adjuntos, es confidencial, privado y de uso exclusivo de los destinatarios a los cuales está dirigido, pudiendo contener información legalmente protegida. Queda prohibida la revisión, divulgación, publicación, modificación, copia, distribución o acción en relación con esta información, por personas o entidades distintas al destinatario. Si usted, por alguna razón, lo ha recibido por error, por favor tenga la amabilidad de eliminarlo de su sistema. Las opiniones contenidas, son exclusivas de su autor, y no representa necesariamente, la opinión de FLENI. http://www.fleni.org.ar
Hi Fernando - Thank you for your kind words. You can indeed try the reinit option as Christofer suggested when one of the pathways looks like a single curve and not like a volume.
However, because some of your other pathways in the screenshots look a bit funky too, I would also check that there was nothing wrong with the gradient directions. An easy way to check that is by looking at the eigenvectors from the tensor fit (dmri/dtifit_V1.nii.gz). Although tracula itself doesn't use tensors, if there's something wrong with the gradient directions, it'll show up in the tensor fit.
If you're not sure how to check, you can upload all tracula-related directories of your subject here and I'll take a look: http://surfer.nmr.mgh.harvard.edu/fswiki/FtpFileExchange
Hope this helps, a.y
On Thu, 18 Oct 2012, Fernando Ventrice wrote:
Dear Freesurfer/Tracula users, Finally I could successfully test Tracula and I think it is amazing what it can do automatically. These are the commands I used:
export FREESURFER_HOME=${HOME}/Programs/freesurfer_510 && source $FREESURFER_HOME/SetUpFreeSurfer.sh source /etc/fsl/fsl.sh export FSLPARALLEL=condor trac-all -prep -c /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/dmrirc_Fernando_Ventrice trac-all -bedp -c /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/dmrirc_Fernando_Ventrice # "trac-all -bedp" gave some errors so I used bedpostx directly: bedpostx /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/20120112_VENTRICE_FERNANDO/dmri trac-all -path -c /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/dmrirc_Fernando_Ventrice freeview -tv /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/20120112_VENTRICE_FERNANDO/dpath/merged_avg33_mni_flt.mgz -v /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/20120112_VENTRICE_FERNANDO/dmri/dtifit_FA.nii.gz &
All finished without errors, but there seems to be a problem with CC Forceps Mayor volume, which gave very small values. Is this OK or I made some mistake? First I tested with some patient and other fiber tracks also gave very small volumes. I attached some pictures and the dmrirc and trac-all log files. Best, Fernando Ventrice
El contenido del presente mensaje y el de sus adjuntos, es confidencial, privado y de uso exclusivo de los destinatarios a los cuales está dirigido, pudiendo contener información legalmente protegida. Queda prohibida la revisión, divulgación, publicación, modificación, copia, distribución o acción en relación con esta información, por personas o entidades distintas al destinatario. Si usted, por alguna razón, lo ha recibido por error, por favor tenga la amabilidad de eliminarlo de su sistema. Las opiniones contenidas, son exclusivas de su autor, y no representa necesariamente, la opinión de FLENI. http://www.fleni.org.ar
Hi Fernando - Thanks for uploading your data. Looking at the primary directions from tensor fit on the data that you sent me (see screenshot), they seem to be off by a 90 degree angle from what you'd expect them. This means that your gradient table is probably off as well.
Hope this helps, a.y
On Thu, 18 Oct 2012, Fernando Ventrice wrote:
Dear Freesurfer/Tracula users, Finally I could successfully test Tracula and I think it is amazing what it can do automatically. These are the commands I used:
export FREESURFER_HOME=${HOME}/Programs/freesurfer_510 && source $FREESURFER_HOME/SetUpFreeSurfer.sh source /etc/fsl/fsl.sh export FSLPARALLEL=condor trac-all -prep -c /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/dmrirc_Fernando_Ventrice trac-all -bedp -c /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/dmrirc_Fernando_Ventrice # "trac-all -bedp" gave some errors so I used bedpostx directly: bedpostx /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/20120112_VENTRICE_FERNANDO/dmri trac-all -path -c /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/dmrirc_Fernando_Ventrice freeview -tv /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/20120112_VENTRICE_FERNANDO/dpath/merged_avg33_mni_flt.mgz -v /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/20120112_VENTRICE_FERNANDO/dmri/dtifit_FA.nii.gz &
All finished without errors, but there seems to be a problem with CC Forceps Mayor volume, which gave very small values. Is this OK or I made some mistake? First I tested with some patient and other fiber tracks also gave very small volumes. I attached some pictures and the dmrirc and trac-all log files. Best, Fernando Ventrice
El contenido del presente mensaje y el de sus adjuntos, es confidencial, privado y de uso exclusivo de los destinatarios a los cuales está dirigido, pudiendo contener información legalmente protegida. Queda prohibida la revisión, divulgación, publicación, modificación, copia, distribución o acción en relación con esta información, por personas o entidades distintas al destinatario. Si usted, por alguna razón, lo ha recibido por error, por favor tenga la amabilidad de eliminarlo de su sistema. Las opiniones contenidas, son exclusivas de su autor, y no representa necesariamente, la opinión de FLENI. http://www.fleni.org.ar
See screenshot now :)
On Tue, 30 Oct 2012, Anastasia Yendiki wrote:
Hi Fernando - Thanks for uploading your data. Looking at the primary directions from tensor fit on the data that you sent me (see screenshot), they seem to be off by a 90 degree angle from what you'd expect them. This means that your gradient table is probably off as well.
Hope this helps, a.y
On Thu, 18 Oct 2012, Fernando Ventrice wrote:
Dear Freesurfer/Tracula users, Finally I could successfully test Tracula and I think it is amazing what it can do automatically. These are the commands I used:
export FREESURFER_HOME=${HOME}/Programs/freesurfer_510 && source $FREESURFER_HOME/SetUpFreeSurfer.sh source /etc/fsl/fsl.sh export FSLPARALLEL=condor trac-all -prep -c /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/dmrirc_Fernando_Ventrice trac-all -bedp -c /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/dmrirc_Fernando_Ventrice # "trac-all -bedp" gave some errors so I used bedpostx directly: bedpostx /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/20120112_VENTRICE_FERNANDO/dmri trac-all -path -c /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/dmrirc_Fernando_Ventrice freeview -tv /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/20120112_VENTRICE_FERNANDO/dpath/merged_avg33_mni_flt.mgz -v /home/labneuroimg/STATION_FLENI/Tensores/Fernando_Ventrice/20120112_VENTRICE_FERNANDO/dmri/dtifit_FA.nii.gz &
All finished without errors, but there seems to be a problem with CC Forceps Mayor volume, which gave very small values. Is this OK or I made some mistake? First I tested with some patient and other fiber tracks also gave very small volumes. I attached some pictures and the dmrirc and trac-all log files. Best, Fernando Ventrice
El contenido del presente mensaje y el de sus adjuntos, es confidencial, privado y de uso exclusivo de los destinatarios a los cuales está dirigido, pudiendo contener información legalmente protegida. Queda prohibida la revisión, divulgación, publicación, modificación, copia, distribución o acción en relación con esta información, por personas o entidades distintas al destinatario. Si usted, por alguna razón, lo ha recibido por error, por favor tenga la amabilidad de eliminarlo de su sistema. Las opiniones contenidas, son exclusivas de su autor, y no representa necesariamente, la opinión de FLENI. http://www.fleni.org.ar
Dear Anastasia, When I open the dtifit_V1 and dtifit_FA with Freeview the vector directions actually are OK, but when I open them with FSLview they are similar to your screen-shot. Here I also send a screen-shot with the comparison between the two softwares in each monitor. Best, Fernando
El contenido del presente mensaje y el de sus adjuntos, es confidencial, privado y de uso exclusivo de los destinatarios a los cuales está dirigido, pudiendo contener información legalmente protegida. Queda prohibida la revisión, divulgación, publicación, modificación, copia, distribución o acción en relación con esta información, por personas o entidades distintas al destinatario. Si usted, por alguna razón, lo ha recibido por error, por favor tenga la amabilidad de eliminarlo de su sistema. Las opiniones contenidas, son exclusivas de su autor, y no representa necesariamente, la opinión de FLENI. http://www.fleni.org.ar
Hi Fernando - I looked at your data and changing the sign of the z coordinate in your gradient table gave me eigenvectors that looked fine in both fslview and freeview. Why you would get eigenvectors that look different in freeview and fslview before flipping z but not after is still a mystery to me, but at this point I'd just try starting clean, changing the sign in the 3rd column of your gradient table, and testing the whole pipeline on a test subject to see if that fixes things. Let me know how it goes!
a.y
Dear Anastasia, When I open the dtifit_V1 and dtifit_FA with Freeview the vector directions actually are OK, but when I open them with FSLview they are similar to your screen-shot. Here I also send a screen-shot with the comparison between the two softwares in each monitor. Best, Fernando
El contenido del presente mensaje y el de sus adjuntos, es confidencial, privado y de uso exclusivo de los destinatarios a los cuales está dirigido, pudiendo contener información legalmente protegida. Queda prohibida la revisión, divulgación, publicación, modificación, copia, distribución o acción en relación con esta información, por personas o entidades distintas al destinatario. Si usted, por alguna razón, lo ha recibido por error, por favor tenga la amabilidad de eliminarlo de su sistema. Las opiniones contenidas, son exclusivas de su autor, y no representa necesariamente, la opinión de FLENI. http://www.fleni.org.ar
freesurfer@nmr.mgh.harvard.edu
-
Anastasia Yendiki -
Fernando Ventrice -
Watson, Christopher