📦 Ingrowth-Segmentation-DSA-Plugin

Ingrowth-Segmentation-DSA-Plugin is a Python package for image segmentation of ingrown structures in microscopy images of Cerebral Aneurysms. It is designed for high-resolution scientific image analysis with support for large formats, deep learning models, and preprocessing pipelines.

⚠️ Note: This plugin is intended for deployment and execution within a Digital Slide Archive (DSA) environment.
It cannot be run directly on a local machine and is designed to be integrated with the DSA system using girder-slicer-cli-web and ctk-cli.

---

🚀 Features

• 🧠 Deep learning segmentation with segmentation-models-pytorch
• 🖼 Handles large microscopy images with opencv-python
• 📊 Rich support for data handling via pandas, openpyxl, and xlrd
• ⚙️ Seamless integration with girder-client, girder-slicer-cli-web, and ctk-cli
• 🧪 Includes preprocessing and augmentation with albumentations

---

🛠 Installation (for development or DSA deployment)

To install the plugin manually (for development or DSA configuration):

git clone https://github.com/SarderLab/Ingrowth-Segmentation-DSA-Plugin.git && \
cd Ingrowth-Segmentation-DSA-Plugin && \
pip install .

Note: This project uses setuptools_scm for versioning. Ensure it is installed if building from source.

---

📁 Repository Structure

Ingrown/
├── __init__.py
├── ... (core segmentation modules)
tests/
setup.py
README.md

---

🧪 Requirements

Major dependencies (automatically installed):

• numpy, pandas, scikit-image, scikit-learn
• opencv-python, Pillow, imageio
• segmentation-models-pytorch, albumentations
• girder-client, girder-slicer-cli-web, ctk-cli
• tqdm, openpyxl, xlrd, joblib

Optional (commented in setup.py):

• torch, torchvision, matplotlib, shapely, dask, tifffile, etc.

---

💻 Usage

This package is designed to be executed as a plugin inside the Digital Slide Archive (DSA) using the slicer CLI interface.
Local usage or direct script execution is not supported.

---

👤 Authors

Sayat Mimar
📧 [email protected]
🧪 Developed at Computational Microscopy Imaging Laboratory, University of Florida

Fatemeh Afsari
📧 [email protected]
🧠 Developed at Computational Microscopy Imaging Laboratory, University of Florida

---

📄 License

Licensed under the Apache License 2.0. See the LICENSE file for more information.

---

📌 Citation

If you use this code for research, please cite this repository or related publications (coming soon).