BrainVISA

Dear Brainvisa users/developpers,

When I am using morphologist the program returns the following error at step 3 of the pipeline (building of brain mask):

------------------------------ !! VIP Error: Not enough memory... ------------------------------ ------------------------------ !! VIP Error: Execution aborted in function '(Vip_alloc.c)VipMalloc'. ------------------------------ ------------------------------ !! VIP Error: Memory pb in realloc ------------------------------ ------------------------------ !! VIP Error: Execution aborted in function 'VipDwindle3DBucket_S16BIT'. ------------------------------
Can't allocate memory - no memory left. This application has requested the Runtime to terminate it in an unusual way. Please contact the application's support team for more information.

However the RAM is far from being full (only 4GB out of 24). I suspect this is happening because 32-bits programs (Brainvisa here) can not allocate more than 2GB on Windows operating system. Can anyone confirm that? Does anyone have the same problem?

Thanks a lot,
Gwenael

Hi,

Indeed, the Windows version is currently only built in 32 bits mode. we have not dealt yet with the task of installing a 64 bits development environment on Windows, compiling all dependency libraries in 64 bits, and testing it. So for now if you need more than 2 GB, you have to use Linux (or potentially a virtual machine containing a 64 bits Linux, that you can run from a Windows OS).
However 2 GB is generally enough to process standard data. What's the resolution of your input image ?

Denis

Hi,

Thanks for your answer.

The resolution of the 3D MRI is 176x240x256. Looks like standard resolution though.

Gwenael

So it's not normal that the process needs so much memory. I guess something went wrong in the previous steps, either in the bias correction or in the histogram analysis steps.
Denis

I tried with other subjects, some worked fine, some had the same error. I have clue: the one I converted from dicom to niftii using dcm2nii (mricron) crashed. The one I converted from dicom to analyze format using a tierce software (cartool) worked fine. So maybe dcm2nii has something wrong.

Gwenael

If the problem comes from dcm2nii, it is strange that the problem only arise at the third step. Have you looked at the intermediate results of steps 1 and 2 ? Could you send us one of your image with problem ?

An error occured at first step and then the memory problem at 3rd step. I attached a screenshot of the log and the MRI data can be downloaded here: http://www.2shared.com/file/a9CUI1jy/mri.html

Gwenael

Hi!

I do have just exactly tha same problem

Do you know, finally, how was it solved??

Thanks in advance