How to replace 0s between 1s with 1s "filling in" meshgrid-derived XYZ matrices?

1 visualizzazione (ultimi 30 giorni)
DICOM ContourData gives points representing organ structures. (Originally a vector for each slice, [x1 y1 z1 x2 y2 z2 ...], I have created three vectors grouping x y z values.)
figure; scatter3(xcontour,ycontour,zcontour,'.')
>> title('PTV1 DICOM Coordinates')
I have another matrix with RTDOSE values. I have created a meshgrid (X,Y,Z) giving the dose's DICOM coordinates and, I think, derived another matrix from it that changes these PTV1 coordinates into 1s with the empty space between them into 0s:
xcontourpointslocation = ismember(X,xcontour);
ycontourpointslocation = ismember(Y,ycontour);
zcontourpointslocation = ismember(Z,zcontour);
So I think now this figure can be thought to represent in "DICOM space" where those 1s are. If I multiply it by the dose matrix, I expect to thereby " extract" the dose at these locations. Yet I obviously want the dose "inside" the structure as well -- i.e., I must convert all those 0s in xcontourpointslocation etc "between the 1s" into 1s to "fill in" the structure.
What MATLAB commands do this elegantly? I have an idea to use the find command with conditioned loops or for loops, but I expect there to be a simpler way, e.g. a command like makesolid(matrix), because this seems straightforward or commonplace.
While hoping you will tell me such a method, I have found some related discussion:
  1. how to fill a column with 1's between two specific values and with 0 otherwise?
  2. [replace with interpolated values] zeros in a [vector]
I will continue working on this problem and post my answer if I solve it before hearing from you.
  3 Commenti
Daniel Bridges
Daniel Bridges il 12 Lug 2016
I want to access all the points corresponding to the interior of that volume and the points on its surface in a corresponding matrix containing dose. My thinking is that if I can use a same-sized meshgrid with all the points on surface & interior being 1, otherwise 0, then I can achieve this through element-wise matrix multiplication.
I am trying to avoid assumptions, since ideally the program should be robust to work with any structure. Moreover, structures may be complicated (e.g. a mixture of concave and convex) as their shape is a function of both anatomy and disease-spread.

Accedi per commentare.

Risposte (1)

KSSV
KSSV il 12 Lug 2016

Categorie

Scopri di più su DICOM Format in Help Center e File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by