Need help on analysing the area for these structures

Hi,

I need suggestions and guidance on how to analyse the area of these structures in an automated manner.


I am trying to do it with cellprofiler but I cannot get it to recognise the structures as objects to measure its area.

Many thanks

Hi @Kmengham,

Welcome to the forum!

Could you explain a bit what you already tried?

I would start with filtering, maybe a gaussian or a median filter, to reduce the noise. Then I would try out all automated threshold methods and see what they can do. If the results is far from acceptable I would probably try to use a blob-detection algorithm.

Cheers

Hi @tibuch,

I have tried playing with the threshold and using the analyze particle function on imagej.

I have also tried playing with smoothing and the findprimaryobject on cellprofiler.

Thank you very much for any help!

Kurtis

If you can post the pipeline you have so far, and the example image(s) that go with it, we could give you a hand. Before the IdentifyPrimaryObjects module in CellProfiler you probably want to add an EnhanceOrSuppress module to enhance the circle shapes.

You might also try ilastik, which works well for non-fluorescence images.

Hi Anne,

Thank you so much for the advice! using that I have made a bit of progress. I currently have used EdgeEnhance as I cannot get the Circle Enhance to work.
coac_Quan.cpproj (407.7 KB)

Using the image I posted above I get to this. However, ideally I would have each circle as its own individual object.

Thank you again for all your help!

Kurtis