Choosing a reconstruction size for vessel segmentation

[Yonni Friedlander]

Hi,

I'm trying to segment the vessels from clinical scans (lung algorithm) that have submillimetre resolution in-plane but 1.25mm slice thickness. My current approach is to go as small as I can while not encountering an "iter max exceeded" message that can appear at the end of the "_stachBlurSpatial" step.

However, I'm not certain what process is reaching its max iterations and if its something that I'm being overly concerned about. 

My the function is being called with something like:

> python /ChestImagingPlatform/Scripts/cip_compute_vessel_particles.py -i ${i%%/}-TLC.nrrd -l partialLungLabelMap.nrrd -r WholeLung --tmpDir /host/ -o particles.vtk --perm -s 0.90

Thanks,

Yonni