Public Repository

Last pushed: 7 months ago
Short Description
This image contains an image analysis pipeline for analysis of whole slide tissue images.
Full Description

This image contains an image analysis pipeline to segment nuclei in whole slide tissue images and compute a number of shape, size, and intensity features for each segmented nucleus.

You can pull the Docker image as follows:

docker pull docker pull sbubmi/pathomics_nucleus:1.0

To use this container, please clone the GitHub repository for the source code:

git clone https://github.com/SBU-BMI/pathomics_analysis.git

This will create a pathomics_analysis folder. You can interact with the Docker image (start a container, segment a region in a whole slide tissue image, and kill and remove the container) using the run_docker_segment.py Python script (located in nucleusSegmentation/script):

Usage: run_docker_segment.py [-h|command <arguments>]

Commands:

start - Start a Docker instance for segmentation.
remove - Kill and remove Docker instance.
segment - Run image analysis pipeline.

Run run_docker_segment.py command for help on command arguments.

run_docker_segment.py start <docker name> [<docker image>]
<docker name> - unique name for the Docker container instance. The Docker container with this <docker name> will be started.
<docker image> - optional name of the Docker image to start.

run_docker_segment.py remove <docker name>
<docker name> - unique name for the Docker container that was provided with the start command. Running Docker container instance will be killed and removed.

run_docker_segment.py segment <docker name> <tissue image file> <output zip file> <other arguments>
<docker name> - the name of the running Docker container.
<image file> - input whole tissue image file.
<zip file> - zip file that will store the result files.
Required arguments:
-s <tile_minx,tile_miny>
-b <tile_width,tile_height>
-d <patch_width,patch_height>
-a <analysis_id: string>
-c <case_id: string>
-p <subject_id: string>
Optional arguments:
-r <otsuRatio>
-w <curvatureWeight>
-l <sizeLowerThld>
-u <sizeUpperThld>
-k <msKernel>
-n <levelsetNumberOfIterations>
-m <mpp>
-e <analysis desc: string>
-v <output level: mask|mask:img|mask:img:overlay>

Docker Pull Command
Owner
sbubmi

Comments (0)