The Francis Crick Institute
Browse
- No file added yet -

Whole slide image analysis of Picrosirius red staining: an automated digital processing workflow

Download (4.63 MB)
poster
posted on 2024-08-13, 10:38 authored by Amy Burley, Priya NarayananPriya Narayanan, Tom Lund, Manuel Salto-Tellez, Alan Melcher, Nick James, Anna Wilkins

Poster presented as part of the Crick BioImage Analysis Symposium 2023.

Capturing brightfield Picrosirius red (PSR) stained images facilitates the visualisation and quantification of collagen matrix features at high resolution. However, staining variation across formalin-fixed paraffin-embedded tumour samples makes highthroughput analysis challenging. Moreover, user-dependent quality-control of whole slide images (WSI) from large sample cohorts is impractical and subjective.

To our knowledge no single tool exists to address the above challenges, as such we developed an automated workflow to process PSR WSI.

The workflow encompasses:

  • artefact correction
  • parameterised quality control
  • stain normalisation
  • one-size-fits-all colour deconvolution

The workflow has been designed to scale in a generic fashion, scale irrespective of the number of pathologist annotations, facilitate the study of tissue heterogeneity and reliably stitch regions of interest for visualisation. Our integrated approach performs at tile level, using effective multi-threading to reduce computational cost.

Permission has been given by authors to upload to Crick Figshare. Copyright remains with the original authors.

History

Usage metrics

    The Francis Crick Institute

    Categories

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC