Get bias field using sitk

I'm trying to use sitk library to clear biased Mr image (MRI).

I'm able to get the corrected image but can't get the bias image.

Set the following parameters:

shrink_factor=1,
mask_image=mask, 
number_of_iterations=100, 
number_of_fitting_levels=4

This is the code I tried

corrected_image = corrector.Execute(image, maskImage)

log_bias_field = corrector.GetLogBiasFieldAsImage(inputImage)
bias_exp = sitk.Cast(sitk.Exp(log_bias_field), sitk.sitkFloat64)
bias = inputImage / bias_exp
bias = sitk.GetArrayFromImage(bias)
bias = bias / bias.max()

But I get this image:

Do you know how to get the bias only?

To get the bias field function you need to decide on what resolution, for the full resolution image:

bias_field_full_resolution = sitk.Exp(corrector.GetLogBiasFieldAsImage(biased_image_full_resolution))

When calling the Execute method the bias field is estimated and the corrected image has the same resolution as the input, which is often a resampled version of the original biased image.

If we actually want to correct the image at the original resolution:

bias_field_full_resolution = sitk.Exp(corrector.GetLogBiasFieldAsImage(biased_image_full_resolution))
corrected_image_full_resolution = biased_image_full_resolution / bias_field_full_resolution