# SlicerSeqSeg Extension

**URL:** <https://discourse.slicer.org/t/slicerseqseg-extension/46707>\
**Category:** Development\
**Tags:** segmentation, extensions-manager, python\
**Created:** [April 10, 2026, 5:29pm UTC](https://discourse.slicer.org/t/slicerseqseg-extension/46707 "2026-04-10T17:29:06Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![numisveinsson](https://avatars.discourse-cdn.com/v4/letter/n/ecccb3/32.png) [@numisveinsson](https://discourse.slicer.org/u/numisveinsson)\
**Post date:** [April 10, 2026, 5:29pm UTC](https://discourse.slicer.org/t/slicerseqseg-extension/46707/1 "2026-04-10T17:29:06Z")

</div>

Hi everyone,

I’m sharing **SeqSeg** , a new **3D Slicer extension** for **seed-based segmentation of tubular/vascular structures** from CT or MR. You place **two fiducial seeds** and a **radius estimate** ; the module runs the **SeqSeg** pipeline (with **nnUNet** ) and loads the segmentation (and optional surface mesh) back into Slicer. Pretrained weights can be pulled from Zenodo via a button in the module UI.

- **Source / documentation:** [https://github.com/numisveinsson/SlicerSeqSeg](https://github.com/numisveinsson/SlicerSeqSeg)

- **Underlying method (Python package):** [https://github.com/numisveinsson/SeqSeg](https://github.com/numisveinsson/SeqSeg)

**Publication**

Sveinsson Cepero, N., Shadden, S.C. _SeqSeg: Learning Local Segments for Automatic Vascular Model Construction._ Ann Biomed Eng **53** , 158–179 (2025). [https://doi.org/10.1007/s10439-024-03611-z](https://doi.org/10.1007/s10439-024-03611-z)

I’m planning to submit the extension to the **Slicer Extensions Index** so it can be installed from the Extension Manager; until then it can be used by adding the module path or building from source as usual.

Feedback, bug reports, and collaboration ideas are welcome—either here or on GitHub issues.

Thanks,  
Numi