[Camino-users] RESTORE question

[Ryan Muetzel]

Dear Philip,

Thanks for the quick response.  Yes, this happens when using a brain mask.  However, there is the "ring" of bright voxels around the edge of the brain in the FA map due to imperfections in the brain extraction. Is it advisable to do a slight "erosion" of the brain mask to eliminate these edge voxels that could be influencing the model?  I've only tested restore on 1 subject, so I will see how frequently it occurs in the rest of my data.

Thanks again for your help!

Best,
Ryan

On Jul 11, 2014, at 6:12 PM, Philip A Cook <cookpa at mail.med.upenn.edu> wrote:

> Hi,
> 
> It can be a bad sign if it happens often. Are you using a brain mask? If not then adding that will likely cut down the number of errors.
> 
> 
> On Jul 11, 2014, at 11:52 AM, Ryan Muetzel <muet0005 at gmail.com> wrote:
> 
>> Hello,
>> 
>> I'm interested in using the camino software to process some of our DTI data, specifically to compute the tensor using the restore method.
>> 
>> I've followed the tutorial on the website for using the restore method.  The algorithm completes and writes out the proper data.  I am using the 2nd method for estimating sigma (from estimatesnr, sigma_mult from the b=0 images).
>> 
>> However, I see this message:
>> 
>> "inverters.RestoreDT_Inversion invert"
>> "Maximum iterations reached without convergence. Final tensor:"
>> 
>> Is this message a problem?  I was initially worried the algorithm was falling back on the OLS/WLS method, but that doesn't appear to be the case (I see that warning message when I reduce the sigma substantially).  The FA maps look very reasonable, and quite similar to the output from FSLs dtifit/wls method.
>> 
>> Thanks in advance for any information! 
>> 
>> All the best,
>> 
>> Ryan
>> 
>> 
>> _______________________________________________
>> Camino-users mailing list
>> Camino-users at www.nitrc.org
>> http://www.nitrc.org/mailman/listinfo/camino-users
>