I have dicom files with slice thickness 2.5 and pixel spacing ['0.703125', '0.703125'] mentioned in the header. I followed a tutorial to convert the raw data into HU with rescale slope and intercept. After processing, my image data is of size 120 x 512 x 512 (3d numpy array). I am kind of confused on the idea of pixel vs voxel. If i visualize the image data by keeping one axis constant in matplotlib, it is in pixels correct ? If so, how to comprehend the idea of spacing and thickness in terms of numpy array. How would i convert the slice thickness to 1.0 mm x 1.0 mm x 1.0 mm. How does the conversion between pixel space and mm happens ?
Asked
Active
Viewed 278 times
0
-
I'm not sure I understand what you want to achieve. You have an image with a voxel size of (0.7mm, 0.7mm, 2.5mm) - you want to transform the voxel size to be (1, 1, 1)? That would imply adding interpolated slices in z direction, and downsampling the images in x/y direction, though I'm not sure this is what you want. – MrBean Bremen Jul 05 '20 at 13:39
-
I believe that is the right interpretation. – Anil Yadav Jul 06 '20 at 00:19