Identifying/counting/measuring geimsa stained nulcei

Hi from Hiroshima.
I’m new to CellProfiler and need some help.
I at hiroshima university am working cell biology related research on mesothelioma cells.
I found some cell profiler pipelines for counting/ measuring DAPI stained cells.
But,
I would like to similar pipeline to work in Giemsa stained cells
I want to identify, count, and measure (relative area) of the Giemsa stained cells on the white background of the images (nuclei are stained blue and cytoplasm pink to orange by eosin).

I have some questions:
Can color images be used? like the one I uploded here GEIMSA_01.JPG
Is it possible to exclude some stain dirt?

I really want to put your package to use as it would greatly facilitate our work in this and other studies. Would also like to become better acquainted with Cellprofiler in general.

Many thanks in advance,
VISHWA JEET AMATYA
Mesothelioma Pathology Research
Department of Pathology,
Hiroshima University.
E-mail: amatya@hiroshima-u.ac.jp


Hello Vishwa,

Please try the attached pipeline. Also note that I started this pipeline from one that Mark Bray in our platform submitted for another user that you could also have found by searching our Forum for giemsa, here: viewtopic.php?f=14&t=1363&hilit=giemsa
I am only saying this for you (and others) as a reminder that this Forum is often a great place for us all to share and discover practical knowledge!

I made notes in the comment boxes at the top of many modules to help guide you in tailoring the pipeline to the rest of your images.

To your questions:

Yes, please see the pipeline. Either use UnmixColors as I did, or ColorToGray

Yes, you may be able to mask out “stain dirt” or other debris. Debris is often very bright, so that you can use the ApplyThreshold module and the “Set pixels below or above the threshold to zero” to set it to zero (assuming that your stain is bright - if not, then invert the brightness using ImageMath).

Let us know if this helps. Good luck!
David
pipeline_DL.cp (7.55 KB)