将3D医学图像分割工具nnInteractive集成到Slicer平台，支持多种交互方式。
SlicerNNInteractive: A 3D Slicer extension for nnInteractive
- 采用客户端-服务器架构，分离计算任务与界面操作。
- 支持点、框、涂鸦、套索等多种交互方式，提升分割效率。
- 兼容性好，可在不同硬件环境下灵活部署，适合临床科研人员。
SlicerNNInteractive 将先进的可提示深度学习3D图像分割框架 nnInteractive 集成至广泛使用的 3D Slicer 平台。该扩展采用客户端-服务器架构，将计算密集型的模型推理与客户端界面解耦，从而消除客户端对高性能硬件的需求，并相比现有 nnInteractive 插件具备更好的操作系统兼容性。在单机上同时运行客户端与服务器端亦可实现，适用于多种部署场景。扩展提供直观用户界面，保留原始框架中所有交互类型（点、边界框、涂鸦、套索提示），并配备完整的快捷键组合，以支持高效工作流。
原文摘要 · Abstract (English)
SlicerNNInteractive integrates nnInteractive, a state-of-the-art promptable deep learning-based framework for 3D image segmentation, into the widely used 3D Slicer platform. Our extension implements a client-server architecture that decouples computationally intensive model inference from the client-side interface. Therefore, SlicerNNInteractive eliminates heavy hardware constraints on the client-side and enables better operating system compatibility than existing plugins for nnInteractive. Running both the client and server-side on a single machine is also possible, offering flexibility across different deployment scenarios. The extension provides an intuitive user interface with all interaction types available in the original framework (point, bounding box, scribble, and lasso prompts), while including a comprehensive set of keyboard shortcuts for efficient workflow.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。