[Insight-users] Bias Field Correction

[Vincent A. Magnotta]

I have been playing around with the ITK bias field correction
application. The resulting image that it generates contains light and
dark bands across the image in the direction that is specified as the
slice direction. Is this a bug or are there some parameter tweaks that
will remove this artifact? I am running the program with the following
parameters:

BiasCorrector --input SerieMR-0002-0001.img --output biascor.img
--class-mean 65 190 220 --class-sigma 10 20 20 --degree 3 --growth 1.001
--shrink 0.5 --max-iteration 3000 --init-step-size 1.001
--use-slab-identification yes --slice-direction 2 --use-log yes
--input-mask mask.img

I have been testing data both on a 3D T1 weighted scan of a brain and a
phantom that contains only a single type of tissue. For the phantom
image, I specified only a single mean and standard deviation. Both
datasets exhibit the same artifact.

Thanks,
Vince

-- 
Assistant Professor
Department of Radiology
0437 JCP
E-mail: vincent-magnotta at uiowa.edu
Phone: 319-356-8255
Fax: 319-353-6275
Website: http://www.radiology.uiowa.edu