Hello Bruce, and Freesurfers,
Sorry for asking an old question, but I really need more confirmation.
I am running recon-all to processing my structural data, but i am not sure do I have to reorient the data or not. The data were acquired in sagital view, and dicom data were converted into nift format using dcm2nii, the transformed nii file (called original image below) are in wrong orienation, in fslview, it's X axis is acturally from anterior to posterior, the Y, Z axises were also wrong. dcm2nii also created a o*.nii image, which is in right orientation. I run recon-all using both images, and the results (volume for each region) are a little bit different, about 0.25%, for example, volume of a brain rigion got with original image is 2000 , but got with o*.nii image is 1995. It's not a big problem, but since they are exactly the same data, why the results are not exactly the same? is it better to use original image or even dicom files or to use reoriented data? the resolution of my data is 1*1*2 mm.
Another related quetion is, if data are left-right fliped, do I have to flip the raw data and run recon-all again or can I just flip left-right of results, i.e., consider volume got for a left region as that of the corresponding right region?
Thank you very much!
2010-02-02
Chunhui Chen _________________
Institute of Cognitive Neuroscience and Learning Beijing Normal University Beijing, China 100875
发件人： Bruce Fischl 发送时间： 2010-01-22 12:38:11 收件人： chenchunhuichina 抄送： 主题： Re: Re: [Freesurfer] Orientation of 001.mgz versus raw dicom data
no, that shouldn't matter On Fri, 22 Jan 2010, chenchunhuichina wrote:
May I ask a related question?
My structural data are acquired in sagital view, do I have to re-orient it before I run recon-all? In fslview, it's X axis is acturally from anterior to posterior.
Thanks!
2010-01-22
chenchunhuichina
发件人： Bruce Fischl 发送时间： 2010-01-22 07:06:58 收件人： yzhangd@artsci.wustl.edu 抄送： freesurfer@nmr.mgh.harvard.edu 主题： Re: [Freesurfer] Orientation of 001.mgz versus raw dicom data
How do you know? It should be in the same space. Bruce
On Jan 21, 2010, at 5:01 PM, yzhangd@artsci.wustl.edu wrote:
I have run the following command to create a 001.mgz file from raw *.dcm data:
recon-all -s <subjid > -i <path to *.dcm data >
The output 001.mgz file is not in the same native space as the original *.dcm data. Could anyone tell me what space the 001.mgz file is in? Does the above command change the orientation (i.e does it rotate/ translate) besides simply converting from .dcm to 001.mgz
thanks a lot
yuning
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Hi Chunhui you should not need to reorient your data. What version are you using? cheers, Bruce
On Tue, 2 Feb 2010, chenchunhuichina wrote:
Hello Bruce, and Freesurfers,
Sorry for asking an old question, but I really need more confirmation.
I am running recon-all to processing my structural data, but i am not sure do I have to reorient the data or not. The data were acquired in sagital view, and dicom data were converted into nift format using dcm2nii, the transformed nii file (called original image below) are in wrong orienation, in fslview, it's X axis is acturally from anterior to posterior, the Y, Z axises were also wrong. dcm2nii also created a o*.nii image, which is in right orientation. I run recon-all using both images, and the results (volume for each region) are a little bit different, about 0.25%, for example, volume of a brain rigion got with original image is 2000 , but got with o*.nii image is 1995. It's not a big problem, but since they are exactly the same data, why the results are not exactly the same? is it better to use original image or even dicom files or to use reoriented data? the resolution of my data is 1*1*2 mm.
Another related quetion is, if data are left-right fliped, do I have to
flip the raw data and run recon-all again or can I just flip left-right of results, i.e., consider volume got for a left region as that of the corresponding right region? >
Thank you very much!
2010-02-02
Chunhui Chen _________________
Institute of Cognitive Neuroscience and Learning Beijing Normal University Beijing, China 100875
·¢¼þÈË£º Bruce Fischl ·¢ËÍʱ¼ä£º 2010-01-22 12:38:11 ÊÕ¼þÈË£º chenchunhuichina ³ËÍ£º Ö÷Ì⣺ Re: Re: [Freesurfer] Orientation of 001.mgz versus raw dicom data
no, that shouldn't matter On Fri, 22 Jan 2010, chenchunhuichina wrote:
May I ask a related question?
My structural data are acquired in sagital view, do I have to re-orient it before I run recon-all? In fslview, it's X axis is acturally from anterior to posterior.
Thanks!
2010-01-22
chenchunhuichina
·¢¼þÈË£º Bruce Fischl ·¢ËÍʱ¼ä£º 2010-01-22 07:06:58 ÊÕ¼þÈË£º yzhangd@artsci.wustl.edu ³ËÍ£º freesurfer@nmr.mgh.harvard.edu Ö÷Ì⣺ Re: [Freesurfer] Orientation of 001.mgz versus raw dicom data
How do you know? It should be in the same space. Bruce
On Jan 21, 2010, at 5:01 PM, yzhangd@artsci.wustl.edu wrote:
I have run the following command to create a 001.mgz file from raw *.dcm data:
recon-all -s <subjid > -i <path to *.dcm data >
The output 001.mgz file is not in the same native space as the original *.dcm data. Could anyone tell me what space the 001.mgz file is in? Does the above command change the orientation (i.e does it rotate/ translate) besides simply converting from .dcm to 001.mgz
thanks a lot
yuning
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu
-
Bruce Fischl -
chenchunhuichina