Hi,
You need to correct for the bias initially (it will mainly affect
the registrations). To do this you need to initially run BET,
then FAST, and then FIRST. I suggest running FIRST with
the run_first_all command, rather than the individual steps,
as it is easier and we have optimised some of the settings
in that script.
All the best,
Mark
On 6 Sep 2010, at 21:23, Nathan Johnson wrote:
> Thanks Matt
>
> If I am going to use FIRST to pull out hippocampal volumes would you
> recommend that I correct for the signal intensity bias before or
> after I BET my structural image?
>
> From: Matt Glasser <[log in to unmask]>
> To: [log in to unmask]
> Sent: Mon, September 6, 2010 4:17:00 PM
> Subject: Re: [FSL] First - Fast Field Inhomogeneity
>
> You can use fast -B <image> to output the B1 bias field corrected
> image.
>
> Peace,
>
> Matt.
>
> From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On
> Behalf Of Nathan Johnson
> Sent: Monday, September 06, 2010 3:13 PM
> To: [log in to unmask]
> Subject: Re: [FSL] First - Fast Field Inhomogeneity
>
> Thanks Matt
>
> I am referring to the B1 field inhomogeneity (not b0) that leads to
> a higher "intensity" at the occipital cortex in my MPRAGE files. I
> am guessing this is what is termed a "signal intensity bias." Does
> FAST have an option that I can execute in the command line to
> correct this?
>
> Cheers
>
> From: Matt Glasser <[log in to unmask]>
> To: [log in to unmask]
> Sent: Mon, September 6, 2010 3:59:01 PM
> Subject: Re: [FSL] First - Fast Field Inhomogeneity
>
> Are you referring to b0 inhomogeneity or signal intensity bias? I
> don't
> think it is common to have significant distortions from b0
> inhomogeneity in
> MPRAGE scans, but FAST can be used to correct for signal intensity
> bias.
>
> Peace,
>
> Matt.
>
> -----Original Message-----
> From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On
> Behalf
> Of Forrest Johnson
> Sent: Monday, September 06, 2010 2:27 PM
> To: [log in to unmask]
> Subject: [FSL] First - Fast Field Inhomogeneity
>
> Hello FSL Experts,
>
> I am currently attempting to pull out hippocampal volumes using
> FIRST and
> would like to correct for field inhomogeneities in my MPRAGE files
> prior to
> running FIRST. That said, I was wondering what tool would be best to
> use for
> this type of correction (FAST?, SUSAN?, or BET?). After sifting
> through the
> forum I have been able to find several replies stating that applying
> SUSAN
> is not necessary, while other replies state that FAST or BET (-B)
> might be
> the best tools to use. Given this information I was wondering if
> someone
> could answer the following questions
>
> 1. What is the best tool to use to correct for field inhomogeneities
> in my
> MPRAGE?
> 2. Should I apply this tool before or after I bet my MPRAGE?
> 3. Would this order of operation be correct?
> a. Correct for field inhomogeneity
> b. Bet
> c. first_flirt
> d. run_first
> e. first_boundary_correct
> f. pull out volumes using fslstats
> g. Apply Sienax to calculate ICV (using VSCALING factor)
>
> Cheers!
>
>
>
|