Dear Chesney,

this looks rather like a memory leak in CAT12 that only occurs ofter processing many subjects. However, I have checked CAT12 for potentially memory leaks in the past and couldn't find larger leaks even for several thousand of subjects in cross-sectional mode. Anyway, this might be different on some computers and is hard to find. Unfortunately the only advice I can give is to split your data and apply the preprocessing to subsets of 80 subjects.

Best,

Christian


>Hi John,
>
>
>
>I have attached the batch file as requested. We’re performing longitudinal volumetry using T1s, using this batch. The error occurs after about 100 scans during the registration.
>
> 
>
>Hope this helps.
>
> 
>
>Thanks,
>
>Chesney
>
 

 

Chesney Craig PhD

 

Research Associate

Department of Exercise and Sport Science

Manchester Metropolitan University Cheshire

Tel: 0161 247 5538

 

 

 

From: John Ashburner [mailto:[log in to unmask]] 
Sent: 06 March 2018 12:35
To: Chesney Craig <[log in to unmask]>
Cc: SPM <[log in to unmask]>
Subject: Re: [SPM] VBM and SBM using CAT 12 : Out of memory error

 

I think we need more information about how you are processing the images, whether they are anatomical images or fMRI, and whether you mean 340 slices or 340 volumes.  If they are slices, have they been concatenated into 3D volumes?  Perhaps you could send the batch job file (which can be saved via SPM's batch system) - providing there's nothing confidential in there.

 

Best regards,

-John

 

 

On 6 March 2018 at 12:01, Chesney E Craig <[log in to unmask]> wrote:

Hello,

I have been having the same 'Out of memory' error as detailed in this thread. However, I am using a 64-bit system (Windows and Matlab 64-bit version), I have 16GB of RAM and I have 700GB of disk space memory available. Considering this, I am very confused about what is causing this issue.

I am trying to pre-process a longitudinal dataset, which includes ~340 images for each time point, with 1 year between each scan.

The only solution in the related threads seems to be to related to 64-bit vs 32-bit systems. Has anyone experienced this problem whilst using a 64-bit system?

Kind regards,
Chesney

Chesney Craig PhD
Research Associate
Dept Exercise and Sport Science
Manchester Metropolitan University




 

--

Prof John Ashburner
Professor of Imaging Science
UCL Institute of Neurology
Queen Square
Wellcome Trust Centre for Neuroimaging
University College London
12 Queen Square, London, WC1N 3BG
E: [log in to unmask]  T: +44 (0)20 3448 4365
http://www.fil.ion.ucl.ac.uk/

"Before acting on this email or opening any attachments you should read the Manchester Metropolitan University email disclaimer available on its website http://www.mmu.ac.uk/emaildisclaimer "