I made 2 labels which included bilateral paracentral lobule and sulcus regions.
Then I extract the values:  
   - MeanCurv
     After mris_anatomical_stats processing, I extracted the MeanCurv by the command,  
  'aparcstats2table --hemi ?h --parc <label> --meas meancurv --subjects <subjects list> --tablefile <output.txt>'

   - ?h.Curv  
      After mris_anatomical_stats processing with the option '-t lh.curv', I extracted the curvature value by the command,
  'aparcstats2table --hemi ?h --parc <label> --meas thickness--subjects <subjects list> --tablefile <output.txt>'

  - H <mean> +- <std>,   H Mean Rectified Surface Integral
    The Mean curvature values were compute by the command,
    'mris_curvature_Stats -l <label> -g -n -m -f white -o <output.txt> <subject> ?h'

I analyzed the each value. The ?h.Curv mean (within the label) has opposite relationship with Z value compared to MeanCurv, while similar relationship with Z value compared to H <mean>.

I heard that the MeanCurv is the integral of the rectified mean curvature of white matter surface (1/r mm), and ?h.curv is the mean curvature of the white matter surface with a Gaussian smoothing kernel applied.
It looks that the MeanCurv and H Mean Rectified Surface Integral has only positive values, while the H <mean> and ?h.curv has positive or negative value.

I wonder whether MeanCurv and ?h.curv has the difference meaning or difference formula.
And I would like to know the meaning of the ?h.curv negative value.
Is H <mean> ( by mris_curvature_Stats ) compute from ?h.curv ?


Thank you.

Han.