CT Image reconstruction from a .mat file.

Hello everyone, I am trying to reconstruct an image from a projection data mat file. It consists of data with 256 angles of projection. This is the program that I wrote to reconstruct the image. I need to interpolate the data to get a smoother image. But when I do that, I get "NaN" so cannot reconstruct. I don't want to use the iradon function.

load proja                 %loads raw data file into the memory
ctim = zeros(64,64);       %specifying the dimensions
image_final = zeros(64,64);
count = 1;
g = input('Number of Projections(max is 256):'); %No of projections to take into account
p = 180/g;
for theta = 0:p:179         %p angles of projection
    proj_angle = proj(count,:);
    [Xi,Yi] = meshgrid(-31:32);
    t = round(Xi*cosd(theta) + Yi*sind(theta))+32;  %formula for backprojection

      t = floor(abs(t));
      t = t.*((t<65).*(t>0)) + 64*(t>64) + 1*(t==0);
  end
  ctim = proj_angle(t);
      image_final = image_final +ctim;
      count = count + 1;
  %display reconstructed image
  figure(1);              
  imagesc(image_final);
  axis('image');
  title('CT reconstructed Image');    %name of the image