Cell Membrane Staining 3D segmentation

Dear all,

We would like to get new ideas/recommendations regarding this 3D image analysis.

We have a timelapse confocal Zstack of cells stained with a membrane dye. Some of the cells are bright, some are dimmer. We need to segment cell volume in order to count the number of vesicles inside the cells over time in a different channel afterwards.

MembraneStainingTimelapse-mini.tif (7.8 MB)

We found these problems and would love to receive some feedback from you (we are not data analysts):

-By far our best option was the Morphological Segmentation pluging after some denoising, but it is still spliting parts of the cells. Is there any pre-processing that could improve the algorithm cell detection and reduce over-segmentation?

-Is there any possibility to fill the inside of the cells to improve segmentation? We tried morphological Closing but because citosol is quite big the output image looks very distorted.

-Any other plugin/tool that would help us with the analysis?

Thanks a lot!

Verónica