they are automatically detected, but sometimes they can be too permissive (or restrictive) in some regions in some acquisitions. If you look through the recon-all.log file in your subject's scripts dir you can see what they are set to (search for mris_make_surfaces), then look at where the surfaces are wrong and see if you think changing some of the thresholds would help. It will say something like:

computing class statistics...
border white:    238722 voxels (1.42%)
border gray      265205 voxels (1.58%)
WM (101.0): 100.4 +- 7.9 [70.0 --> 110.0]
GM (76.0) : 75.1 +- 10.0 [30.0 --> 110.0]
setting MIN_GRAY_AT_WHITE_BORDER to 61.0 (was 70)
setting MAX_BORDER_WHITE to 113.9 (was 105)
setting MIN_BORDER_WHITE to 71.0 (was 85)
setting MAX_CSF to 50.9 (was 40)
setting MAX_GRAY to 98.1 (was 95)
setting MAX_GRAY_AT_CSF_BORDER to 66.0 (was 75)
setting MIN_GRAY_AT_CSF_BORDER to 40.9 (was 40)


and you can change any of them with expert options (look in the recon-all --help).

cheers
Bruce


On Mon, 17 Feb 2014, amirhossein manzouri wrote:

Thanks Bruce for the prompt reply. Would you please explain how can I
control mris_make_surfaces parameters, is this a part of recon-all process?

Best regards,
Amirhossein Manzouri 

 



On Mon, Feb 17, 2014 at 3:35 PM, Bruce Fischl <fischl@nmr.mgh.harvard.edu>
wrote:
      Hi Amirhossein

      it's hard to tell without looking at the data, but you could try
      using the expert options to change the allowable intensity
      ranges for the different surfaces (e.g. if the white is too far
      out, don't allow mris_make_surfaces to put the gray/white border
      at intensities that are that dark)

      cheers
      Bruce


      On Mon, 17 Feb 2014, amirhossein manzouri wrote:

            Hi all, 
            There is incorrect pial surface lines (red and
            yellow) in the right
            hemisphere of some of my subjects. The white matter
            and grey matter are very
            close regarding intensity and this can be seen in
            just a few slices and then
            it gets correct. I tried control points but
            regarding the low intensity they
            can be effective. I wonder if there is a way to
            correct this, or if they may
            affect the qdec analysis between group afterwards a
            lot.(since they are
            present in one or few slices and create thick GM in
            those slices!)

            Best regards,
            Amirhossein Manzouri 

             





The information in this e-mail is intended only for the person to whom
it is
addressed. If you believe this e-mail was sent to you in error and the
e-mail
contains patient information, please contact the Partners Compliance
HelpLine at
http://www.partners.org/complianceline . If the e-mail was sent to you
in error
but does not contain patient information, please contact the sender
and properly
dispose of the e-mail.