As I said, you can write out the bias corrected image as default.
This is the (bias) corrected image, which should be more homogeneous.
It's name starts with prefix m.
The uncorrected image is the original image, which contains the
intensity inhomogeneity you want to correct.
The segmentation function estimates a bias field for this correction and
applies it to the input image.
Michael.
Am 14.11.2012 23:04, schrieb Daniel Sheltraw:
> What are the "corrected" and "uncorrected images"?
>
>
>> Hi Daniel,
>>
>> in the segmentation function you can specify to write out the bias
>> corrected image (default). The filename starts with prefix m. You can
>> calculate the bias field by dividing the corrected image (ms*.nii for
>> MPRAGE and mf*.nii for EPI) voxelwise by the uncorrected image. You can
>> apply this bias field just by voxelwise multiplikation with the other
>> EPIs.
>>
>> Hope that helps
>> Michael.
>>
>> Am 14.11.2012 21:15, schrieb Daniel Sheltraw:
>>> Hi
>>>
>>> A 2012 NeuroImage paper by Bandettini et al titled "Effects of image
>>> contrast on functional MRI image registration" states the following:
>>>
>>>
>>> "Intensity inhomogeneity corrected datasets (both EPI and MPRAGE) were
>>> computed using the segmentation function available in SPM8
>>> (www.fil.ion.ucl.ac.uk/spm/software). For steady-state EPI data,
>>> intensity
>>> correction was computed based on the bias maps computed for the sixth
>>> volume
>>> of each angle and each subject separately."
>>>
>>> The authors are trying to correct for image inhomogeneity due primarily
>>> to
>>> receive field and transmit field inhomogeneity.
>>>
>>> Does anybody know which SPM function in particular the authors used and
>>> what
>>> the algorithm is for it?
>>>
>>> Thank you,
>>> Daniel
>>>
>>
>> --
>> <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
>> Dr. Michael Erb
>> Dept. Biomedical Magnetic Resonance
>> University Hospital Tübingen
>> Hoppe-Seyler-Str. 3
>> 72076 Tübingen, Germany
>> Tel.: +49(0)7071/2987753 priv. +49(0)7071/61559
>> Fax.: +49(0)7071/294371
>> e-mail: <[log in to unmask]>
>> www:
>> http://www.medizin.uni-tuebingen.de/Forschung/Kliniken/Radiologie/Biomedizinische+Magnetresonanz.html
>> <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
>>
>
--
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
Dr. Michael Erb
Dept. Biomedical Magnetic Resonance
University Hospital Tübingen
Hoppe-Seyler-Str. 3
72076 Tübingen, Germany
Tel.: +49(0)7071/2987753 priv. +49(0)7071/61559
Fax.: +49(0)7071/294371
e-mail: <[log in to unmask]>
www:
http://www.medizin.uni-tuebingen.de/Forschung/Kliniken/Radiologie/Biomedizinische+Magnetresonanz.html
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
|