could you explain to me how i would be able to use the dir function, to do a loop through a directory and all of the subfolders in the directory for files that have the following format.

i have several subfolders in my directory. specifically ones that correspond to a date (ie. 20091011), and in each of those folders, there are several subfolders corresponding to a specific image set. In each of those folders are about 47 files, and I need to process the most recent/the last file in those folders

I00000XX (XX = is a series of numbers corresponding to the file name)…. there is no file extension. (its a DICOM file).

The subfolders are as follows:

ImageData (contains 1 folder) > MDX1 (contains one folder) > image (contains several folders, each corresponding to a date) > 20091002 (contains several folders, each corresponding to a image set) > E0000012 > all of the files, of which the last file is the file I need.

I want to create a list of sorts of all of these files (ie, path).

Then I want to run an executable on all of these files, to get an output that i can save in matrix form

thanks much!

%"matchedPixelUV" is an Nx6 array of matching pixel pairs between video
%frames, in this order: frame# pixel_u pixel_v frame# pixel_u pixel_v.

%Set up an array of unique points from all points in matchedPixelUV.
points = [matchedPixelUV(:,1:3);matchedPixelUV(:,4:6)];
points = unique(points,'rows');

%Use Delaunay Tessellation to get the indices to the unique point list from
%the original matchedPixelUV.  The k1 and k2 vectors are the matching node
%pairs, which will be used to build the adjacency matrix.
T = delaunayn(points);
[k1 d1] = dsearchn(points,T,matchedPixelUV(:,1:3));
[k2 d2] = dsearchn(points,T,matchedPixelUV(:,4:6));
display(['distance checks should be zero: d1=' num2str(sum(d1)) ', d2=' num2str(sum(d1))])
clear T d1 d2 matchedPixelUV

%Now build a sparse adjacency matrix.  Make sure it is square by
%temporarily putting ones in the upper left and lower right of the
%diagonal, then remove them.
A = sparse([1; k1; size(points,1)],[1; k2; size(points,1)],1);
clear k1 k2
A(1,1) = 0;
A(size(points,1),size(points,1))=0;

%Get the transpose of the matrix and add it back to it.  This is because
%[k1 k2] is the "forward" connection between nodes, i.e. [1 2; 1 3; 2 3; 3 4].
%However, it does not have the "return" paths [2 1; 3 1; 3 2; 4 3].
A = A+A';
% spy(A)

%Add the identity matrix so we have ones along the diagonal.
I = speye(size(A));
A = A+I;
clear I

%Use dmperm (Dulmage-Mendelsohn decomposition) to compute the connected
%components, see http://blogs.mathworks.com/steve/2007/03/20/connected-component-labeling-part-3/
%for example.
[p,q,r] = dmperm(A);

Sincerely,
Tommy