{"name":"NuPatch3D","display_name":"NuPatch-3D","visibility":"public","icon":null,"categories":[],"schema_version":"0.1.0","on_activate":null,"on_deactivate":null,"contributions":{"commands":[{"id":"NuPatch3D.make_widget","title":"Open NuPatch3D Tool","python_name":"NuPatch3D._widget:NuPatch3DWidget","short_title":null,"category":null,"icon":null,"enablement":null}],"readers":null,"writers":null,"widgets":[{"command":"NuPatch3D.make_widget","display_name":"NuPatch3D","autogenerate":false,"default_dock_area":"right","tabify":false}],"sample_data":null,"themes":null,"menus":{},"submenus":null,"keybindings":null,"configurations":{}},"package_metadata":{"metadata_version":"2.4","name":"NuPatch3D","version":"1.0.0b1","dynamic":null,"platform":null,"supported_platform":null,"summary":"A napari plugin for 3D cell label correction using GMM and Deep Learning.","description":"# NuPatch3D Tool\n\n[← Repository overview](../README.md)\n\n**NuPatch3D Tool** is a [napari](https://napari.org/) plugin for post-processing 3D nuclear segmentation results. It provides interactive tools to correct under-segmentation, over-segmentation, missing nuclei, and boundary inaccuracies.\n\nLabel merging also supports removal of false-positive nuclear instances by assigning their labels to the background.\n\n![NuPatch3D correction functions and representative annotation results](../assets/figures/NuPatch3D_CoreFunction_example.png)\n\n**Figure 1.** Core correction functions of NuPatch3D and representative human GBM and mouse pancreas annotations, comparing initial Cellpose segmentations with refined results.\n\n## Features\n\n- **Local editing** — Extract and refine individual cells or neighborhoods in an isolated bounding box; changes are written back to the global segmentation only after you Commit\n- **Under-segmentation split** — Separate merged cells using plane-based, GMM auto, manual center modes\n- **Over-segmentation merge** — Merge fragmented labels or delete spurious ones\n- **Boundary refinement** — Adjust cell boundaries based on local image intensity\n- **Missing cell detection** — Add omitted cells via interactive seed points and watershed\n- **Full undo/redo & JSON logging** — Up to 10-step undo/redo with traceable, reproducible editing workflow; supports recovery from JSON logs\n- **Seamless napari integration** — Works with native napari brush, eraser, and fill tools; all operations are recorded in the unified edit history\n\n## Requirements\n\n- **OS:** Windows 10 or macOS 26 (Tahoe) and above, as listed in the existing user manual.\n- **RAM:** 16 GB or more.\n- **Python:** >=3.10; Python 3.11 is recommended for the installation environment.\n- **NumPy:** >=2.0,<3; the plugin uses `np.asarray(..., copy=True)`, introduced in NumPy 2.0.\n- **napari:** >=0.6,<0.7; other versions may have compatibility issues.\n\nThese requirements apply to the NuPatch3D plugin. Benchmark segmentation methods require separate, method-specific environments. The version bounds reflect API requirements, not a fully tested environment lockfile.\n\n## Installation\n\nInstall napari and the plugin in your Python environment:\n\n```bash\nconda create -n NuPatch3D python=3.11\nconda activate NuPatch3D\npip install NuPatch3D\n```\n\nFor local development:\n\n```bash\ngit clone https://github.com/QuKunLab/NuPatch3D.git\ncd \"NuPatch3D/NuPatch3D-tool\"\npip install -e .\n```\n\nLaunch napari with `napari`, then open **Plugins → NuPatch3D (NuPatch-3D)**. Load your microscopy image and instance-label segmentation, and select the corresponding layers in the plugin panel.\n\n## Documentation\n\n- [English user manual](https://qukunlab.github.io/NuPatch3D/en/)\n- [中文用户手册](https://qukunlab.github.io/NuPatch3D/)\n- [Installation and setup](../docs/en/environment.md)\n- [Loading images and segmentation labels](../docs/en/loading.md)\n- [Saving results](../docs/en/save.md) and [recovering edits](../docs/en/restore.md)\n\n## Related modules\n\n[NuPatch3D Dataset](../NuPatch3D-dataset/README.md) · [Benchmark](../Benchmark/README.md)\n","description_content_type":"text/markdown","keywords":null,"home_page":null,"download_url":null,"author":null,"author_email":"\"J. Wang\" <jinbowang@mail.ustc.edu.cn>","maintainer":null,"maintainer_email":null,"license":"MIT","classifier":["Framework :: napari","Programming Language :: Python :: 3"],"requires_dist":["napari[all]<0.7,>=0.6","numpy<3,>=2.0","qtpy","scikit-image","scikit-learn","scipy","pandas","torch","tqdm"],"requires_python":">=3.10","requires_external":null,"project_url":null,"provides_extra":null,"provides_dist":null,"obsoletes_dist":null},"npe1_shim":false}