NVIDIA Clara Deploy SDK User Guide

12.16. 3D Image Processing Pipeline using FastIO

The 3D Image Processing pipeline is one of the reference pipelines provided with Clara Deploy SDK. It accepts a volume image in MetaImage format, and optionally accept parameters for cropping. The output is the cropped volume image and the image is published to the Render Server so that it can be viewed on the web browser.

The 3D Image Processing pipeline using FastIO is defined in the Clara Deploy pipeline definition language. This pipeline utilizes built-in reference containers to construct the following operator:

The following is the details of pipeline definition, with comments describing each operator’s functions as well as input and output.

Copy
Copied!
            

api-version: 0.4.0 name: fastio-itk-pipeline parameters: CROP_START_X: -1 CROP_START_Y: -1 CROP_START_Z: -1 CROP_SIZE_X: -1 CROP_SIZE_Y: -1 CROP_SIZE_Z: -1 operators: - name: read-volume description: Read a volume file into the shared memory container: image: clara/fastio-itk tag: latest command: ["/bin/bash", "-c", "python-u/app/main.pyread_volume"] input: - path: /input output: - path: /output requests: memory: 8192 - name: filter-volume description: Read a volume from the shared memory and pass to filters container: image: clara/fastio-itk tag: latest command: ["/bin/bash", "-c", "python-u/app/main.pyfilter_volume"] variables: CROP_START_X: ${{CROP_START_X}} CROP_START_Y: ${{CROP_START_Y}} CROP_START_Z: ${{CROP_START_Z}} CROP_SIZE_X: ${{CROP_SIZE_X}} CROP_SIZE_Y: ${{CROP_SIZE_Y}} CROP_SIZE_Z: ${{CROP_SIZE_Z}} input: - from: read-volume path: /input output: - path: /output requests: memory: 8192 - name: write-volume description: Write the volume from the shared memory to the file container: image: clara/fastio-itk tag: latest command: ["/bin/bash", "-c", "python-u/app/main.pywrite_volume"] input: - from: filter-volume path: /input - path: /config # user-provided input payload including config_render.json file output: - path: /output requests: memory: 8192 - name: register-volume-images-for-rendering description: Register volume images, MHD format, for rendering. container: image: clara/register-results tag: latest command: ["python", "register.py", "--agent", "renderserver"] input: - from: write-volume path: /input

Please refer to the Run Reference Pipelines using Local Input Files in the How to run a Reference Pipeline section to learn how to register a pipeline and execute the pipeline using local input files.

Example)

Copy
Copied!
            

mkdir -p ~/.clara/pipelines cd ~/.clara/pipelines # Download pipeline clara pull pipeline clara_fastio_itk_pipeline cd clara_fastio_itk_pipeline # Unzip test input data unzip app_fastio_itk-input_v1.zip -d input # Create a pipeline clara create pipeline -p fastio-itk-pipeline.yaml clara create jobs -p <PIPELINE ID> -n <JOB NAME> -f <INPUT PATH> # Start a job with a parameter `CROP_START_X=230` and `CROP_SIZE_Z=300` clara start job -j <JOB ID> -a CROP_START_X=230 -a CROP_SIZE_Z=300

Input requires a folder containing the following files:

  • .mhd (with .raw) - Input MetaIO file
  • config_render.json - Configuration for Render Server

Cropped volume on Render Server

  • Go to the Clara RenderServer UI using a web browser: The URL is: <IP of the machine>:8080
  • You should see a dataset with a name that includes the name of the job you specified and the operator name (e.g., fastio-itk-test-write-volume)
  • Clicking the dataset would show the rendered image on the screen
© Copyright 2018-2020, NVIDIA Corporation. All rights reserved. Last updated on Jun 28, 2023.