5 November 2026
We have added nnInteractive Agent to SlicerNNInteractive, bringing segmentation through plain-language requests into 3D Slicer. The integration builds on nnInteractive Agent by Fabian Isensee and colleagues. You describe what you want segmented, and an AI agent locates the structures, places prompts for nnInteractive, inspects the results, and refines them in response to your feedback.
The agent takes over the sequence of interactions that a human would otherwise perform manually. For example, a request such as “segment every rib individually” gives it a task to work through using the image and the segmentation tools. Claude Code and Codex are supported, alongside configurable model servers. The underlying segmentation still comes from nnInteractive; the agent decides how to use it to carry out the request.
This is the second demonstration from the repository, showing the agent integration in Slicer:
The conversation and the images inspected by the agent appear in the module, while completed objects are added to a Slicer segmentation. You can review the result in the familiar slice and 3D views, ask for changes, or send an existing segmentation to the agent for further work. This creates a workflow that combines text instructions with visual inspection and manual editing.
The agent can use image renderings and associated information when communicating with its configured model provider. Choose a provider or locally hosted model appropriate for your data. Installation and configuration details are in the SlicerNNInteractive README.
Please use this when referring to the post or publication:
C. de Vente, K.V. Venkadesh, A. Lasso, B. van Ginneken and C.I. Sánchez. "SlicerNNInteractive: A 3D Slicer extension for nnInteractive", Journal of Open Source Software, 2026;11(123):9003.
F. Isensee, C. Ulrich Harsy, M. Rokuss, B. Hamm, P. Vollmuth and K.H. Maier-Hein. "nnInteractive Agent: General-Purpose AI Agents as 3D Image Annotators", arXiv preprint arXiv:2610.31080, 2026.
F. Isensee, M. Rokuss, L. Krämer, S. Dinkelacker, A. Ravindran, F. Stritzke, B. Hamm, T. Wald, M. Langenberg, C. Ulrich, J. Deissler, R. Floca and K. Maier-Hein. "nnInteractive: Redefining 3D Promptable Segmentation", arXiv preprint arXiv:2503.08373, 2025.
C. de Vente, K.V. Venkadesh, A. Lasso, B. van Ginneken and C.I. Sánchez. "SlicerNNInteractive: A 3D Slicer extension for nnInteractive", Journal of Open Source Software, 2026;11(123):9003.
@article{devente2026joss,
title={SlicerNNInteractive: A 3D Slicer extension for nnInteractive},
author={de Vente, Coen and Venkadesh, Kiran Vaidhya and Lasso, Andras and van Ginneken, Bram and Sánchez, Clara I.},
journal={Journal of Open Source Software},
year={2026},
volume={11},
number={123},
pages={9003},
doi={10.21105/joss.09003}
}
F. Isensee, C. Ulrich Harsy, M. Rokuss, B. Hamm, P. Vollmuth and K.H. Maier-Hein. "nnInteractive Agent: General-Purpose AI Agents as 3D Image Annotators", arXiv preprint arXiv:2610.31080, 2026.
Copy
@Preprint{isensee2026agent,
title={nnInteractive Agent: General-Purpose AI Agents as 3D Image Annotators},
author={Isensee, Fabian and Ulrich Harsy, Constantin and Rokuss, Maximilian and Hamm, Benjamin and Vollmuth, Philipp and Maier-Hein, Klaus H.},
journal={arXiv preprint arXiv:2610.31080},
year={2026},
url={https://arxiv.org/abs/2610.31080},
website_reference_only={true}
}
F. Isensee, M. Rokuss, L. Krämer, S. Dinkelacker, A. Ravindran, F. Stritzke, B. Hamm, T. Wald, M. Langenberg, C. Ulrich, J. Deissler, R. Floca and K. Maier-Hein. "nnInteractive: Redefining 3D Promptable Segmentation", arXiv preprint arXiv:2503.08373, 2025.
@Preprint{isensee2025nninteractive,
title={nnInteractive: Redefining 3D Promptable Segmentation},
author={Isensee, Fabian and Rokuss, Maximilian and Krämer, Lars and Dinkelacker, Stefan and Ravindran, Ashis and Stritzke, Florian and Hamm, Benjamin and Wald, Tassilo and Langenberg, Moritz and Ulrich, Constantin and Deissler, Jonathan and Floca, Ralf and Maier-Hein, Klaus},
journal={arXiv preprint arXiv:2503.08373},
year={2025},
url={https://arxiv.org/abs/2503.08373},
website_reference_only={true}
}
Google Scholar
LinkedIn
GitHub