I read 3D volume whose size 240x240x155 and I want to read slice by slice and wants to apply feature extraction techniques on them How I can read one slice than other?
1 view (last 30 days)
Show older comments
Javaid Iqbal
on 2 Jun 2018
Commented: Javaid Iqbal
on 8 Jun 2018
Hello Please, may you help me to read the 3D volume in MATLAB?
v=niftiread('Brats17_2013_2_1_flair.nii.gz');
figure, imshow(v(:,:,85),[]);
%how to read slice#1 than slice#2 and so on
0 Comments
Accepted Answer
Walter Roberson
on 2 Jun 2018
for slicenumber = 1 : size(v,3)
this_slice = v(:,:,slicenumber);
this_result = YourFeatureExtractionCallGoesHere(this_slice);
all_results{slicenumber} = this_result;
end
Sometimes other data structures would be more appropriate than a cell array: it depends on what data type is returned by YourFeatureExtractionCallGoesHere
3 Comments
Walter Roberson
on 8 Jun 2018
Yes, but it is unlikely that would be efficient if you are still building the data structure.
If you have a cell array in which the entries are all the same size, then you can use something like
cat(3, all_results{:})
to get a 3 dimensional array of the component parts.
But if that was your aim, then better would be to just store them in 3D in the first place:
numslices = size(v,3);
for slicenumber = 1 : numslices
this_slice = v(:,:,slicenumber);
this_result = YourFeatureExtractionCallGoesHere(this_slice);
all_results(:,:,slicenumber) = this_result;
if slicenumber == 1 && numslices ~= 1
all_results(end,end,numslices) = 0;
end
end
The if logic there takes the 2d array all_results of unknown size and extends it to have numslices panes in an efficient way
More Answers (0)
See Also
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!