[Insight-developers] Sled's N3 Bias field correction in ITK

[Karthik Krishnan]

Great.

Thanks for your contribution.

It seems to work great. I've been trying it on prostate MRI. Occassionally
(one in 4 times), there seems to be some sort of ill-conditioning in the
result. I'm not sure where it stems from. The maximum value in the dataset
goes to "inf". I'll try to post the data, so this becomes reproducible.

Other than that, the only minor change that I made to it, (I'll include that
in the commit) is progress reporting of iteration events. I am using this
from an end user application and one would go nuts without progress :)

I'll use the updated version on the Insight Journal.

Thanks again
--
karthik

On Tue, Jun 9, 2009 at 12:38 PM, Nick Tustison <ntustison at wustl.edu> wrote:

> Hi Karthik,
> As a follow-up to yesterday's email, I looked at the N3 code one more time
> to see about possible changes.  There were two small changes that I made
> which were updated in revision 3 of the code which has been posted to the
> IJ.
>
> 1. I changed the variable BiasFieldControlPointLattice to
> LogBiasFieldControlPointLattice which is more accurate.  These changes were
> propagated to the rest of the code including the test file.  All 6 tests
> continue to pass.
> 2. I had started to write a comment in the .h file which I had left as a
> sentence fragment which has since been fixed.
>
> Since I didn't see any commits of the N3 code to the Review/ directory, I
> thought that this would be okay.
>
> Nick
>
> On Mon, Jun 8, 2009 at 10:01 AM, Karthik Krishnan <
> karthik.krishnan at kitware.com> wrote:
>
>> Nick:
>>
>> Having needed a non-parametric bias field correction method, I plan to add
>> your contribution from the Insight Journal on nonparametric Bias field
>> correction using Sled's N3 to ITK's Review. I was wondering if you had any
>> updates/changes since the version posted on the Insight journal.
>>
>> Please let me know.
>>
>> Thanks
>> --
>> karthik
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/mailman/private/insight-developers/attachments/20090609/b129d2d9/attachment.htm>