Finding the right parameters and options

Dear all,

I would like to analyze nuclear staining intensities in fluorescence images. However, I do not seem to be able to find the right modules and parameters to do this job in various images reliably. Perhaps a more experienced user could help me setting up a working (or roughly working) pipeline to do the job.
I think the images need illumination correction and the right set of values to faithfully segment the nuclei. If this is a problem: I would be just as happy to provide seeds for nucleus segmentation manually somehow? Could one edit the final result by e.g. clicking on wrongly identified cells and exclude them from the final analysis and vice versa clicking on dubious objects in the results table and see the corresponding cell in the image to decide whether to exclude it or not?
The resulting measurements should reflect e.g. ‘greenness’ and ‘redness’ of nuclei (mean and integrated density).

Any help would be greatly appreciated.


Hi Matthias,

I’m attaching a pipeline which should help. Basically, what needs tweaking is the smoothing filter size and the maxima separation distance settings in IdentifyPrimaryObjects. Check out the module notes at the top of the settings panel for more details.

2011_03_17.cp (6.64 KB)