Hi Bo
The mask is a binary label indicating the brain or intracranial
portion of the image. You can create one with FSL:
bet image bet -n -m
The mask image will be called bet_mask.nii.gz.
My scaling method relies on information only from within the image
itself. It does not use any external reference. This should normally
suffice to compensate for scanner differences, as long as both produce
T1-weighted images. Within the brain, any scanner should detect low
signal for CSF and high signal for white matter. It's a different
matter if gadolinium contrast has been used for one or both
acquisitions.
Hope that helps
Rolf
On 11 June 2012 00:23, Suchada Tantisatirapong <[log in to unmask]> wrote:
> Hi Rolf,
>
> Thank you for the script. Could you please explain how to create a mask? This mask should be used as a standard template?
>
> In common, do people use T1W or T2W as reference image? From the FSL registration pratice, T1w is used as a reference image. What should be the criterion to select the reference image, e.g. resolution, FOV? I guess that I should use image which have greater FOV as a reference image.
>
> Many thanks
> Bo
>
> ________________________________________
> From: FSL - FMRIB's Software Library [[log in to unmask]] On Behalf Of Rolf A. Heckemann [[log in to unmask]]
> Sent: 10 June 2012 16:59
> To: [log in to unmask]
> Subject: Re: [FSL] intensity normalization
>
> Hi Bo
>
> For what it's worth, I attach a script for normalizing the intensity
> range of an image based on the robust range within a mask (which you
> need to have prepared).
>
> Usage:
>
> scale-generic.sh image mask output
>
> Hope that helps
>
> Rolf
>
>
>
> On 10 June 2012 14:50, Suchada Tantisatirapong <[log in to unmask]> wrote:
>> Hi,
>>
>>
>>
>> I’m trying to normalize intensity of structural MRI images because these
>> images are acquired from different scanner, generating with different
>> intensity range.
>>
>>
>>
>> I use a simple math like:
>>
>>
>>
>> New_intensity = constant* (old_intensity – min_intensity)/(max_intensity –
>> min_intensity)
>>
>>
>>
>> Does FSL offer any more advance technique for normalizing image intensity?
>>
>>
>>
>> Many thanks for your reply
>>
>> Bo
>
>
>
> --
> Rolf A Heckemann, MD PhD
> Médecin chercheur, Fondation Neurodis
> CERMEP - Imagerie du Vivant
> Hôpital Neurologique Pierre Wertheimer
> 59 Boulevard Pinel
> 69003 Lyon
> France
>
> 2nd affiliation: Honorary Fellow, Imperial College London
--
Rolf A Heckemann, MD PhD
Médecin chercheur, Fondation Neurodis
CERMEP - Imagerie du Vivant
Hôpital Neurologique Pierre Wertheimer
59 Boulevard Pinel
69003 Lyon
France
2nd affiliation: Honorary Fellow, Imperial College London
|