Just wondering if anyone knows if I am going wrong anywhere, I have hundreds of CT scan images, similar to the one below. I have been able to generate a segmentation/reconstructed CAD geometry (STL File) from these images.
However, when I look at the size of the STL model, it is not comparable to the sample that I have CT images for, in terms of size. So I am wondering if it is at all possible to use the known the image separation distance (~6 um) to then generate an STL file with the correct dimensions? Or rather use the image separation distance and image size (x by y number of pixels)?
Probaby the imported image has incorrect physical scale, hence all downstream objects carried this over.
Can you check what the image spacing is reported for your volumes and that they are correct?
For anyone with the same issue, the following, is the process required to use generate a 3D CAD/STL file segmentation that can be exported using a 2D image stack (multiple CT scan images), with a known pixel spacing.
Download 3D Slicer and have it installed.
At the top of 3D Slicer, use View —> Extension Manager and search (in the top right) for SliceMorph.
Download and install and accept all of the required toolboxes (initially 5 and an extra 1 will popup).
Modules: SlicerMorph (bottom of dropdown) —> Input and Output —> ImageStacks. Or search for imagestacks using the magnifying glass.
Using the Input Files — > Filename pattern: Select the first image in the sequence.
In the spacing: update this to the correct Image Pixel Size, in my case this was ~6um but I input this as 0.006 mm.
For a high quality CAD model, ensure that full model is selected in the Quality section.
Click Load Files.
Modules: Segmentation Editor —> Add.
Threshold (select appropriate). REMEMBER TO CLICK APPLY.
In my case, I used the Island tab to remove small islands. Again remember to click apply.
Modules: Segmentations.
Representations —> Click Create next to Closed Surface.
