Skip to content

A WIPP Plugin for inferencing using StarDist 2D pretrained models

License

Notifications You must be signed in to change notification settings

usnistgov/WIPP-StarDist-inference-plugin

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

WIPP StarDist inference plugin

Statements of purpose and maturity

The purpose of this work is to create a WIPP plugin based on the StarDist 2D Object detection package.

Description of the repository contents

  • src: contains the source Python code
  • Dockerfile
  • plugin.json WIPP plugin manifest

Technical installation instructions, including operating system or software dependencies

The code is written in Python 3 (tested on version 3.10) and leverages the tensorflow and stardist python packages.

Installation (optional if using the pre-built Docker image)

Build Python Virtual Environment

conda create --name stardist python=3.10
conda install grpcio
pip install tensorflow stardist imagecodecs
conda activate stardist

Build the Docker image

docker build . -t wipp/wipp-stardist-inference-plugin:0.0.1

Execution

Pre-trained model choices from the StarDist package are: 2D_versatile_fluo, 2D_paper_dsb2018 and 2D_versatile_he

Run the Python code

From this directory:

python ./src/stardist-inference.py \
--inputImages ./sample-data/images \
--output ./sample-data/outputs
--pretrainedModel 2D_versatile_fluo

Run the Docker image

From this directory, assuming the images to process are in a folder "sample-data/images":

docker run -v "$PWD"/sample-data:/data \
wipp/wipp-stardist-inference-plugin:0.0.1 \
--inputImages /data/images \
--output /data/outputs \
--pretrainedModel 2D_versatile_fluo

-v: mounts a volume/folder from your machine inside of the Docker container

Run the WIPP plugin

- register the plugin.json in a deployed WIPP instance - see https://github.com/usnistgov/WIPP
- upload input images as WIPP image collection
- create a workflow by adding one step called stardist-inference
- run and monitor the workflow execution
- download resulting WIPP image colection

Additional Information

Contact information

  • WIPP team, ITL NIST, Software and System Division, Information Systems Group
  • Contact email address at NIST: wipp-team@nist.gov

Related Material

Citation:

Uwe Schmidt, Martin Weigert, Coleman Broaddus, and Gene Myers. Cell Detection with Star-convex Polygons. International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), Granada, Spain, September 2018.

About

A WIPP Plugin for inferencing using StarDist 2D pretrained models

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published