Rotated Slices of 3D volumes

Hello, I have a 3D volume of voxels, this volume only contains integers 1,2 and 3. If I took slices parallel to principal planes ( x-y, x-z, y-z) the slice.CData only contains integers 1,2 and 3, which is correct. But if I used a rotated slice, slice.CData contains numbers inbetween and many NaNs. which may be because the slice does not cut exactly through the voxel. So my question is, is there a method to make slices that gives me the integer of the closest voxel which the slice cut through. To Clarify this , we can run the example given in Matlab documentation:

 [x,y,z] = meshgrid(-2:.2:2,-2:.25:2,-2:.16:2);
v = x.*exp(-x.^2-y.^2-z.^2);
figure
colormap hsv
for k = 0
   hsp = surf(linspace(-2,2,20),linspace(-2,2,20),...
      zeros(20) + k);
   rotate(hsp,[1,-1,1],30)
   xd = hsp.XData;
   yd = hsp.YData;
   zd = hsp.ZData;
   delete(hsp)

   slice(x,y,z,v,[-2,2],2,-2)
   hold on
   S = slice(x,y,z,v,xd,yd,zd)
   hold off
   view(-5,10)
   axis([-2.5 2.5 -2 2 -2 4])
   drawnow
end
A = S.CData;
unique(A)

Answer 1

[Rik]

You can use round on the output data, although I don't see where your data would have the values you mention. The code below should help.

A = S.CData;
A(A<1)=1;
A(A>3)=3;
A=round(A);
S.CData=A;

[Ahmed Zankoor]

Thank you for your reply. This is a good solution and I tried it and it seems working. In the attached files, the slice in the 3D space (1.fig) contains numbers between 1,2 and 3 besides NaNs (CData1.mat), the imagesc() of rounded CData (CData2) is shown in (3.fig), notice that NaNs are treated as zeros !!). I noticed that these NaNs are obtained when the slice is out the boundary of the 3D volume, and I suppose the inbetween numbers are obtained since the axes scale of the slice is not the same as the axes scales of the 3D volume (row, column, block of voxels) yet I am not sure how to adjust both axes. Thanks

[Rik]

Glad to be of help. I don't really understand if you still have a question or not, so if you have a follow-up question, please clarify what it is in a comment.

[Ahmed Zankoor]

Hello rik, it has been a while since this discussion, but I just wanted to say that using 'nearest' option in slicing can be useful to avoid the previous interpolation. S = slice(V,Xp,Yp,Zp,'nearest');