OPEN_NONPLANAR contours are not supported yet, but support may be added.

What do these structures represent?

They represent applicators. (Applicators are used in the context of brachy therapy.)

I will definitely fix the crash, that shouldn't occur.

Andras, do you think it would be difficult to add support for open contours? If it's represented the same way, then the reading part is done, but we need to represent them differently. Ribbons maybe?

We plan to add support for reading/writing HDR brachy catheter paths soon.

For visual representation tube models would work well. However, users would need to access the actual curve (centerline). I guess we can store both the curve and the tube model in a segmentation object, similarly to the raw closed contour and closed surface for usual structures.

Corresponding issue in SlicerRT that tracks the commits
https://www.assembla.com/spaces/slicerrt/tickets/786-add-support-for-open-contours

@Richard Mueller

1. What tube radius would you think reasonable for the loaded catheter paths? It will be a conversion parameter for sure, the question is what should be the default? It can be a constant radius, it can depend on the slice thickness of the other planar contours in the RTSS, or it can come from other modalities in the study (I don't know if it's specified i the plan for example).
2. Can you please attach the related image and plan so that we can add an automatic test for the whole study?
   Thanks!

I fixed the crash, so it can load at least the closed planar contours correctly. I will add the open contour support soon.

Usually the radius of the applicators is between about 1.35 and 2.5 millimeters. I do not know what other planning systems export but we do currently not export the applicator radius in any DICOM file.

For me it is absolutely sufficient that Slcier does not crash when open nonplanar structures are in the Struct files. It is not necessary to display the applicator paths. Anyhow, if they were displayed that would be a nice feature.

In brachytherapy we are actually more interested in the dwell positions then the applicators themselves since the dose calculation is based on the dwell positions and their direction vectors.

I have uploaded the study. See zip archives SagiNova1 to SagiNova8.

Thank you for the data! If you update SlicerRT this afternoon, it shouldn't crash any more.
Loading the planned paths will be necessary eventually anyway, for reconstructing the catheters from tracking data during treatment.

If you could verify that there is no crash any more, then I think we can close this issue, as it is about the crash. If you're interested in the open contour loading, then you can monitor the assembla ticket mentioned above.
Thanks!

The error is solved. Thanks!