Pycroglia

A Python toolkit for quantitative 3D morphology analysis of cells from fluorescence microscopy images.

Originally based on the MATLAB tool CellSelect-3DMorph, Pycroglia reconstructs individual cells voxel-by-voxel and extracts quantitative morphological descriptors. It is built with a robust, extensible architecture and supports both a PyQt6 graphical interface and a library mode for automated/scripted workflows.

---

Key Features at a Glance

• Multi-format I/O — reads TIFF and LSM image stacks.

• Interactive filtering — per-slice Otsu threshold, small-object removal, morphological erosion.

• Cell segmentation — connected-component labeling with optional GMM-based cell splitting.

• 3D skeletonization — Lee-Kashyap-Chu thinning (slimskel3d) or scikit-image fallback.

• Morphological metrics — volume, centroid, territorial volume, branch lengths, ramification index.

• Parallel computation — Qt thread pool and multiprocessing backends behind a unified Pool façade, orchestrated by a DAG-based scheduler.

• Rich export — Excel (XLSX), JSON, and 3D geometry (OBJ / PLY / VTP / VTK / VTI).

• Wizard GUI — step-by-step workflow: File Selection → Filters → Segmentation → Cell Selection → Results Dashboard.

---

User Guide

• Getting Started
  • Requirements
  • Installation
  • Launching the GUI
  • Library Mode
  • Next Steps
• Workflow and UI Guide
  • 1. File Selection
  • 2. Filter Editor
  • 3. Segmentation Editor
  • 4. Cell Selection
  • 5. Results Dashboard
• System Architecture
  • Core Module (pycroglia.core)
  • UI Module (pycroglia.ui)
• Project
  • Issue association
  • Creating Issues
  • Branch naming conventions
  • Commit Message Guidelines
  • Pull request guidelines
• Python
  • Docstrings
  • Linter and formatter

API Reference

• pycroglia
  • pycroglia package
    • Subpackages
    • Module contents