Help on Wrong labelling

Okay, so I really have some problems with tracking my Foci.

First I’ll give you a generelle idea of what I’m hopping to get.

So the data I have is images of a number of Foci within a 2 different Nucleus. My data are first Z-stackes, and then time runs.
So each of my subfolder contains 16 images of the same with different height of the Nucleus, and each subfold is then a different time of the same.
The Foci disappear/appear as a function of Z(Goes out of focus), and as a function of time. What We actually want to measure is how intensity decreases for some Foci and how
some “New” Foci appear and start to increase in intensity.

Since some appear and disappear I use LAP 2nd Phase tracking(since this seems to be the ones I should choose). I can upload my whole Pipeline it that would help.
But here are my trouble with tracking visuallized.

The first image, I don’t really find all foci that are there.
https://lh6.googleusercontent.com/_EyNzrR3V9R8/TdEjp-7tSwI/AAAAAAAAABM/LNsJXqn6_1I/cellHelp_1.png
Then I use tracking, which of course just give me this one:
https://lh5.googleusercontent.com/_EyNzrR3V9R8/TdEjp7aY6BI/AAAAAAAAABQ/H4ofmgc3lvQ/Cellhelp_2.png
Now the next run is giving me this for identifying Foci:
https://lh6.googleusercontent.com/_EyNzrR3V9R8/TdEjdhvCD6I/AAAAAAAAABE/-eohwQTUfFg/cellhelp_3.png
But now the tracking is very wrong: Foci 1, is not Foci 1 but 6 and so on.
https://lh5.googleusercontent.com/_EyNzrR3V9R8/TdEjj0DUgcI/AAAAAAAAABI/jcJ80JLrgJg/cellHelp_4.png

Hello,

CellProfiler does not currently have the ability to do 3D segmentation, which unfortunately sounds like what you would really want. But, there may be a workaround. Perhaps the thing you really want to do is first run a pipeline to turn each Z stack into a single image, using the MakeProjection module (the pipeline would be: LoadImages, MakeProjection, SaveImages). Once those ‘flattened’ images are produced and saved, then you could run a second pipeline where you use LoadImages, IdentifyPrimaryObjects, and TrackObjects. If in this pipeline you are still having trouble finding foci accurately, you might check the “Speckle Counting” example pipeline on the Examples page of the main CellProfiler website (cellprofiler.org/examples.shtml). There are certainly many tricks for find objects when they are against a hazy background fluorescence in the cell - you might check other forum posts related to speckles to find more ideas on this.

Best wishes,
Anne