https://github.com/ajinkya-kulkarni/pyorganoidnet
This repository provides StarDist and CellPose models, meticulously trained on a large dataset of Pancreatic Ductal Adenocarcinoma organoids co-cultured with immune cells. Pre-print available at https://www.biorxiv.org/content/10.1101/2024.02.12.580032v1. Demo application available at https://segmentorganoids.streamlit.app/
https://github.com/ajinkya-kulkarni/pyorganoidnet
cellpose deep-learning organoid-segmentation organoids stardist
Last synced: 5 days ago
JSON representation
This repository provides StarDist and CellPose models, meticulously trained on a large dataset of Pancreatic Ductal Adenocarcinoma organoids co-cultured with immune cells. Pre-print available at https://www.biorxiv.org/content/10.1101/2024.02.12.580032v1. Demo application available at https://segmentorganoids.streamlit.app/
- Host: GitHub
- URL: https://github.com/ajinkya-kulkarni/pyorganoidnet
- Owner: ajinkya-kulkarni
- License: gpl-3.0
- Created: 2023-09-09T08:54:49.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2024-07-05T07:38:56.000Z (almost 2 years ago)
- Last Synced: 2025-09-04T19:50:01.718Z (9 months ago)
- Topics: cellpose, deep-learning, organoid-segmentation, organoids, stardist
- Language: Jupyter Notebook
- Homepage: https://zenodo.org/records/10643410
- Size: 60.9 MB
- Stars: 2
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
- Citation: CITATION.cff
Awesome Lists containing this project
README
[](https://segmentorganoids.streamlit.app/)
[](https://www.gnu.org/licenses/gpl-3.0)
[](https://zenodo.org/records/10643410)


# OrganoIDNetData: Pancreatic Ductal Adenocarcinoma Organoid Dataset
## Introduction
Welcome to the OrganoIDNetData repository. This public dataset is a significant step forward in cancer research, particularly in the study of Pancreatic Ductal Adenocarcinoma (PDAC). It comprises phase-contrast images of murine and patient-derived tumor organoids co-cultured with immune cells. With 190 images and 33,906 organoids, OrganoIDNetData serves as a potential benchmark for organoid segmentation models in oncological research.
The pre-print based on this work can be found [here](https://www.biorxiv.org/content/10.1101/2024.02.12.580032v1.full.pdf).
## Dataset Overview
- **Type of Cancer:** Pancreatic Ductal Adenocarcinoma
- **Images:** 190 phase-contrast images
- **Organoids Count:** 33,906
- **Culturing:** Co-cultured with immune cells
- **Focus:** Tumor organoids
## Objective
The primary objective of OrganoIDNetData is to address the challenges in organoid research, particularly:
- Efficient and reliable segmentation of organoid images
- Quantification of organoid growth, regression, and response to treatments
- Prediction of organoid system behaviors
## Usage
This dataset is intended for use in developing and testing algorithms for:
- Object detection and segmentation in organoid images
- Machine learning models in oncology research
- Benchmarking against other organoid segmentation models
## Contributing
We welcome contributions to OrganoIDNetData! If you have suggestions or improvements, please fork the repository and submit a pull request.