Fitting surface to 3D data from a CT stack

I have about 100 slices of CT images of size 512 by 512. I segmented the bone from it and got a 3D matrix. Now I want to identify how the left side of the bone model is different from the right. So, I am trying to fit a surface to it and analyze the surface. I am unable to fit a surface to the 3D model I have. Most of the surface fitting I found had no more than 3 slices of data.

[Matt J]

What surface model did you use?

[kishore]

I tried using cftool but realized that I need to have x,y & z variables separately! So I could not do it. Should I resize the matrix before I try fitting a surface?

[Matt J]

What is the model equation for the surface?

Answer 1

[Sean de Wolski]

Try using lsqcurvefit in the Optimization Toolbox or NonLinearModel.fit in the Statistics Toolbox. These allow for an arbitrary number of dimensions.

[kishore]

I only have Image Processing & Curve Fitting Toolboxes. I tried to download lsqcurvefit source code but it did not work. Is there any other way to do it.