I’m currently using 3D Slicer for manual segmentation of cardiac CTs to quantify epicardial adipose tissue. The CTs that I have are high resolution at a slice thickness of 0.5mm or less (some as small as 0.15mm). Within 3D Slicer, is there a way to “merge” segments so that the thickness is only 1mm (or perhaps even 2mm?). I know this will diminish the accuracy of the measurements to some extent, but it may be worth it to us for the time saved in analysis. I appreciate any advice on how to do this.
Yes, you can set the geometry of the segmentation independent of the background (it’s set to the background by default). Use the segmentation geometry button next to the master volume selector.
https://slicer.readthedocs.io/en/latest/user_guide/image_segmentation.html#basic-concepts
Thank you for this quick reply. Would you be able to provide additional details of where I need to go to adjust slice thickness? I apologize in advance, but I’m a novice at this and would appreciate any additional information. I’m using the latest version of Slicer.
Hmm, good question. I thought you could just edit this in the segmentation geometry dialog, but I can’t seem to enable that mode easily. @lassoan or others?
You may also choose to segment just 1 out of every 5-10 slices and use “Fill between slices” effect to create a full, high-resolution segmentation by smoothly interpolating between the segmented slices. You can insert additional slices anywhere where you find that the interpolated segmentation is not accurate enough.
Thank you both, this is a great option. The issue I am running into now is that after “filling between slices,” the quantification of adipose tissue volume appears to be a lot lower than I would expect, approximately by a factor of 10. Because I manually segmented every 10th slice, I am wondering if the quantification is not capturing the slices that were segmented using the “fill between slices” feature. Is there a setting that I need to change?
It seems that you have not clicked “Apply” in “Fill between slices” effect to write the preview of interpolation results into the segmentation.