MATLAB Answers

how can cropped subvolume containing all voxels that have intensity in the interval [-100,200]HU from 3D ct images

1 view (last 30 days)
how can crooped subvolume containing all voxels that have intensity in the interval [-100,200]HU from 3D ct images

Accepted Answer

Walter Roberson
Walter Roberson on 4 Aug 2016
mask = YourVoxels >= -100 & YourVoxels <= 200;
any1 = any(mask,1);
any12 = any(any1, 2);
start_pane = find( any12, 1, 'first');
end_pane = find( any12, 1, 'last');
any13 = any(any1, 3);
start_col = find( any13, 1, 'first');
end_col = find( any13, 1, 'last');
any23 = any(any( mask, 2), 3);
start_row = find( any23, 1, 'first');
end_row = find( any23, 1, 'last');
sub_volume = YourVoxels(start_row : end_row, start_col : end_col, start_pane : end_pane );
Alternately,
mask = YourVoxels >= -100 & YourVoxels <= 200;
rinfo = regionprops( double(mask), 'BoundingBox');
bb = rinfo.BoundingBox;
sub_volume = YourVoxels(bb(1) : bb(1) + bb(4) - 1, bb(2) : bb(2) + bb(5) - 1, bb(3) : bb(3) + bb(6) - 1);
Or if you want that more verbosely:
mask = YourVoxels >= -100 & YourVoxels <= 200;
rinfo = regionprops( double(mask), 'BoundingBox');
bb = rinfo.BoundingBox;
start_row = bb(1);
end_row = bb(1) + bb(4) - 1;
start_col = bb(2);
end_col = bb(2) + bb(5) - 1;
start_pane = bb(3);
end_pane = bb(3) + bb(6) - 1;
sub_volume = YourVoxels(start_row : end_row, start_col : end_col, start_pane : end_pane );
  2 Comments
Walter Roberson
Walter Roberson on 4 Aug 2016
My first version works without changes.
The second version should be
mask = YourVoxels >= -100 & YourVoxels <= 200;
rinfo = regionprops( double(mask), 'BoundingBox');
bb = ceil(rinfo.BoundingBox);
sub_volume = YourVoxels(bb(2) : bb(2) + bb(5) - 1, bb(1) : bb(1) + bb(4) - 1, bb(3) : bb(3) + bb(6) - 1);
The image you posted is not a 3D CT image: it is a 2D slice of a CT image that has had its values shifted and scaled, and then it has been blurred (by saving as JPEG.) The YourVoxels array that you should be using is the data that you get from dicomread()

Sign in to comment.

More Answers (0)

Community Treasure Hunt

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

Start Hunting!

Translated by