Does anyone has a protocol for the quantification of Sirius red-stained tissues with ImageJ or also any other tool?

Hi. I am a novice to histology as well as the analysis of histological images and I have used Sirius Red staining to stain for collagen in my lung samples. Can anyone please help me with a protocol or SOP with Image J (or any other tool) which I can use to quantify the area of my slide covered by fibrotic tissue (often accompanied by excessive and disordered collagen depostiion)? I have attached one of my images below in which I have tried to identify what I think are areas of fibrosis and areas of normal collagen cocentration. I want to quantify areas of fibrosis and I also want to avoid quantifying areas with normally high collagen content such as areas of the large airways.

Sirius red will stain collagen in red. I am also open to proposals of other tools that can help me too.

Thank you in advance for your kind help.


This is a great question of me as well. I am trying to figure out collagen staining and quantification on muscles. I used to apply an approach with CellProfiler but found it’s not ideal, which could distinguish tendon from fibrosis. Please keep me updated if you have any solutions.

I think this was a duplicate of

We do this fairly regulary with heart sections, but the usual output is just percentage area of the tissue or area of interests (say blood vessel) that is sirrius red positive. Easily done in imageJ/Fiji as a threhsold measurment or scripted to be more automated.

Several other options for you to work with these type of images woudl be to look at machine learining like Ilastik or Weka in Fiji that let you train the system for detection of the type of tissue. You could train for normal and fibrotic to get refined areas for simple analysis or take them further for more advanced analysis.

If you have access to polarisation microsocpy you can capture really nice images that are a bit easy to quatitate. See here for some good examples (

A more advanced apprach is using second harmonic geneation. This requires access to a multiphoton microsocpe system with the right detectors on it but this allows you to look at the native collagen without the need for a stain. If you have both forward and back path detectors you can start to look at collagen types as shown in this paper ( If you don’t need that level just looking at the forward proegated signal works too as shown here (

1 Like

If you are looking into other imaging methods, fluorescence is usually fairly straightforward for quantification as well.

@cbdai2000 for QuPath, I recently added some tutorials that may be relevant:

Thank you Pete for your great help! I am trying out today.


1 Like

Thank you all for your instant help. I am trying to detect fibrosis - collagen as red in the PSR staining, without tendon or blood vessels. I think this need machine learning to tell the difference. Let me play around first.