Prostate cancer classification workflow
Version 1

Workflow Type: Common Workflow Language
Stable

Prostate cancer classification workflow

This workflow segments tissue regions and classifies prostate cancer on H&E whole slide images, using AI. It consists of three steps:

  1. low-resolution tissue segmentation to select areas for further processing;

  2. high-resolution tissue segmentation to refine borders - it uses step 1 as input;

  3. high-resolution normal/cancer classification - it uses step 1 as input.

Click and drag the diagram to pan, double click or use the controls to zoom.

Inputs

ID Name Description Type
gpu n/a n/a
  • int
slide n/a n/a
  • File
tissue-high-batch-size n/a n/a
  • int
tissue-high-chunk-size n/a n/a
  • int
tissue-high-filter n/a n/a
  • string
tissue-high-label n/a n/a
  • string
tissue-high-level n/a n/a
  • int
tissue-low-batch-size n/a n/a
  • int
tissue-low-chunk-size n/a n/a
  • int
tissue-low-label n/a n/a
  • string
tissue-low-level n/a n/a
  • int
tumor-batch-size n/a n/a
  • int
tumor-chunk-size n/a n/a
  • int
tumor-filter n/a n/a
  • string
tumor-label n/a n/a
  • string
tumor-level n/a n/a
  • int

Steps

ID Name Description
classify-tumor n/a n/a
extract-tissue-high n/a n/a
extract-tissue-low n/a n/a

Outputs

ID Name Description Type
tissue n/a n/a
  • File
tumor n/a n/a
  • File

Version History

Version 1 (earliest) Created 2nd Apr 2025 at 15:36 by Simone Leo

No revision comments

Open master 1b23832
help Creators and Submitter
Creator
  • Mauro Del Rio
Submitter
License
Activity

Views: 41   Downloads: 11

Created: 2nd Apr 2025 at 15:36

help Tags

This item has not yet been tagged.

help Attributions

None

Total size: 11.6 KB
Powered by
(v.1.16.0)
Copyright © 2008 - 2024 The University of Manchester and HITS gGmbH