Adjust Label after resampling medical volume
1 回表示 (過去 30 日間)
古いコメントを表示
I resampled a medical volume (nifti) following these instructions:
ctFile=medicalVolume('E:\git\maskRCNN\norm\LD\case_00167.nii.gz');
targetVoxelSize=[3.11 1.59 1.59];
ratios = targetVoxelSize ./ ctFile.VoxelSpacing;
origSize = size(ctFile.Voxels);
newSize = round(origSize ./ ratios);
origRef = ctFile.VolumeGeometry;
origMapping = intrinsicToWorldMapping(origRef);
tform = origMapping.A;
newMapping4by4 = tform.* [ratios([2 1 3]) 1];
newMapping = affinetform3d(newMapping4by4);
newRef = medicalref3d(newSize,newMapping);
newRef = orient(newRef,origRef.PatientCoordinateSystem);
newCTFile = resample(ctFile,newRef,Method='linear');
write(newCTFile,"E:\git\maskRCNN\norm\LD\case_00167");
ctFile is an image with Voxels 512x512,103 , taken from the kits19 challenge (https://github.com/neheller/kits19/tree/master/data)
However, the label (also a nifti file) that marked a organ on the original image (ctFile), does not mark the same area on the new, resampled image (newCTFile). I assume I need to tranform the label too. How do I do that?
1 件のコメント
cr
2022 年 11 月 14 日
Please consider providing more information like data sizes, code etc for community to be able to understand and possibly answer.
採用された回答
Harsha
2022 年 11 月 18 日
Hi Odo Luo
Consider the Labels Nifti files also as a medical volume and apply all the transformations(that are performed on the CT file) as given in the link mentioned by you.
その他の回答 (0 件)
参考
カテゴリ
Help Center および File Exchange で Biomedical Imaging についてさらに検索
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!