search

vistalab / vistasoft

VISTASOFT is the main software repository of the VISTA lab at Stanford University.
http://vistalab.stanford.edu
148 stars 141 forks source link

Issue with nifti2ROI #313

Closed hhal closed 4 years ago

hhal commented 5 years ago

Hi all,

I'm having an issue with the nifti2ROI script.

I've defined a lesion mask in a stroke patient, and want to load this lesion as a ROI in mrVista. Lesion is in 'RAS'. The mask aligns with the t1 thas has been used in the mrVista analysis.

I'm using the nifti2ROI script to import this ROI and that has nifti2mrVistaAnat implemented which should re-orient the nifti to mrVista Anatomy convention.

The script runs errorless, however, the lesion ROI still seems not to be oriented correctly. In fact, it falls completely outside the brain.

When, for example, I'm using the same script to install the probability atlas of Wang (as described here) I don't run into any problems.

Any clue what when wrong?

JWinawer commented 5 years ago

Does the lesion ROI file have the identical header to the T1 anatomy?

On Tue, Apr 16, 2019 at 6:18 AM hhal notifications@github.com wrote:

Hi all,

I'm having an issue with the nifti2ROI script.

I've defined a lesion mask in a stroke patient, and want to load this lesion as a ROI in mrVista. Lesion is in 'RAS'. The mask aligns with the t1 thas has been used in the mrVista analysis.

I'm using the nifti2ROI script to import this ROI and that has nifti2mrVistaAnat implemented which should re-orient the nifti to mrVista Anatomy convention.

The script runs errorless, however, the lesion ROI still seems not to be oriented correctly. In fact, it falls completely outside the brain.

When, for example, I'm using the same script to install the probability atlas of Wang (as described here) I don't run into any problems.

Any clue what when wrong?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_vistalab_vistasoft_issues_313&d=DwMCaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=W4u2U_DhoNoJOD0UdsYQiqH56Bg3zFfOtNa0RCuuEIs&m=eu3WOsk_U2beibW7eOm-PTUtI85oH2cGihYbJUwq5hI&s=KoBB0mCBxtd4qtl2ttO7TcpiyXxFk0k5v7sKpRYo4E4&e=, or mute the thread https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_notifications_unsubscribe-2Dauth_ACBX3pc8eZJVh8ZD9JieGhmAViECNmW7ks5vhaOMgaJpZM4cx4QI&d=DwMCaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=W4u2U_DhoNoJOD0UdsYQiqH56Bg3zFfOtNa0RCuuEIs&m=eu3WOsk_U2beibW7eOm-PTUtI85oH2cGihYbJUwq5hI&s=pBGM4sFjWt9lkJh3_Ui2Dc9EHOsVU5SyOfsT6Wng-Y8&e= .

-- Jonathan Winawer Assistant Professor of Psychology and Neural Science

New York University 6 Washington Place New York, NY, 10003 (212) 998-7922 (phone) (212) 995-4018 (fax) jonathan.winawer@nyu.edu http://psych.nyu.edu/winawer/

hhal commented 5 years ago

Hi Jonathan,

Thanks for the swift reply.

Just wanted to update on this issue. In the meantime I found the cause of the misalignment. And yes, indeed by checking the header information I found out that the qoffset_x/y/z values differed between the two niftis!

Best,

On 16 Apr 2019, at 14:57, Jonathan Winawer notifications@github.com wrote:

Does the lesion ROI file have the identical header to the T1 anatomy?

On Tue, Apr 16, 2019 at 6:18 AM hhal notifications@github.com wrote:

Hi all,

I'm having an issue with the nifti2ROI script.

I've defined a lesion mask in a stroke patient, and want to load this lesion as a ROI in mrVista. Lesion is in 'RAS'. The mask aligns with the t1 thas has been used in the mrVista analysis.

I'm using the nifti2ROI script to import this ROI and that has nifti2mrVistaAnat implemented which should re-orient the nifti to mrVista Anatomy convention.

The script runs errorless, however, the lesion ROI still seems not to be oriented correctly. In fact, it falls completely outside the brain.

When, for example, I'm using the same script to install the probability atlas of Wang (as described here) I don't run into any problems.

Any clue what when wrong?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_vistalab_vistasoft_issues_313&d=DwMCaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=W4u2U_DhoNoJOD0UdsYQiqH56Bg3zFfOtNa0RCuuEIs&m=eu3WOsk_U2beibW7eOm-PTUtI85oH2cGihYbJUwq5hI&s=KoBB0mCBxtd4qtl2ttO7TcpiyXxFk0k5v7sKpRYo4E4&e=, or mute the thread https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_notifications_unsubscribe-2Dauth_ACBX3pc8eZJVh8ZD9JieGhmAViECNmW7ks5vhaOMgaJpZM4cx4QI&d=DwMCaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=W4u2U_DhoNoJOD0UdsYQiqH56Bg3zFfOtNa0RCuuEIs&m=eu3WOsk_U2beibW7eOm-PTUtI85oH2cGihYbJUwq5hI&s=pBGM4sFjWt9lkJh3_Ui2Dc9EHOsVU5SyOfsT6Wng-Y8&e= .

-- Jonathan Winawer Assistant Professor of Psychology and Neural Science

New York University 6 Washington Place New York, NY, 10003 (212) 998-7922 (phone) (212) 995-4018 (fax) jonathan.winawer@nyu.edu http://psych.nyu.edu/winawer/ — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/vistalab/vistasoft/issues/313#issuecomment-483648112, or mute the thread https://github.com/notifications/unsubscribe-auth/Ak5iawE6Y65eT3z50_fAV7SR-8aMtM7Xks5vhcjGgaJpZM4cx4QI.

JWinawer commented 4 years ago

This issue looks resolved. Please re-open if not.