Hello everyone,
I am working on a DICOM dataset with three volumes (axial, sagittal, and coronal) of a brain MRI study. Each volume is clear and high-quality for its respective orientation (e.g., axial cuts in the axial volume), but the other orientations appear very pixelated.
My goal is to create a single merged volume that combines the high-quality slices from each orientation (axial, sagittal, and coronal) into one coherent dataset that I can use for segmentation and further analysis.
I have attempted the following:
- Volume Registration: Using the General Registration (BRAINS) module, I aligned the volumes. However, the output remains pixelated in non-primary orientations.
- Resampling: Adjusting B-Spline grid size, percentage of samples, and interpolation mode. The results are still not satisfactory.
- Merging Volumes: I looked for ways to combine the clear slices from each volume into a single node but couldn’t find an effective solution.
Here are my questions:
- Is there a recommended method to merge slices from different volumes while retaining their original quality?
- Are there specific settings I should tweak in the registration or resampling process to improve the results?
- Should I use a different approach entirely to achieve my goal?
Thank you for your guidance!
