Possible to track multiple cells with temporally variable morphologies?



This is my first visit to this site, so please let me first introduce myself. My name is David Schwarz and my labs discipline is immuno-oncology. Our current interest is macrophage development and function, and we recently acquired an Etaluma LS560 to enable real-time video microscopy. The first projects have been cytokine-driven in vitro macrophage differentiation from human monocytes. We have quite decent data sets now, and we would like to evaluate those data so that we can assign measurable indices to the biologies we have observed.

Our monocytes begin life as a somewhat round cell, but over the course of several days they will change shape - sometimes quite dramatically - as they migrate around the plate. My question - is there software beyond trackmate that will allow us to isolate a subset of cells in the field (5-10), assign them each a unique color, and then track their various movements over time? I’m looking to track not only overall movement, but also changes in shape and polarity.

We are at best new to ImageJ, and at worst the annoying new user asking questions many times addressed in the forum - but not knowing the best key words to find the answers we seek. We’re not looking for someone to hold our hand through this, only a simple “yes, there is software for this”, or “you aren’t the first to want to do this, but at present there isn’t a consolidated software package - I’m afraid you’ll have to do it manually”.

Thanks for reading and any assistance you can offer.


Hi David
Although I don’t do imaging over course of several days, I am familiar with growing and differentiation of human monocytes to macrophages. The difference is that I image these differentiated macrophages .

Anyway, I can imagine there would be dramatic changes in cell morphology. Do you use human monocyte cell lines or are these isolated from human blood? Do the cells have any fluorescent tags or is it mainly brightfield images?

Post a few images of different time frames and the kinds of images you have acquired (brightfield, and/or fluorescence?). Are these just one picture every time or Z stacks?
Also, how many images do you generate over one experiment? Many people here may not have a background in biology, but are well equipped to offer you assistance if you post the images.




Dear Pradeep,

Thank you for your reply. We are working with purified human monocytes isolated from peripheral blood. We collect a phase-contrasted bright-field image every five minutes, and typically have ~1,200 images when the study is complete. These are not stacked in the Z-plane. Indeed the morphological changes are extensive - I have attached two (TIFF) images titled start and finish - I believe they are self-explanatory.

I’m grateful for your response. The lack of any other responses leads me to believe that there isn’t anything available “out of the box” that can facilitate the types of evaluations we wish to undertake. But I’ll gladly take any further direction that can be offered.

Best regards,

Edit - seems the server didn’t like my TIFF images. Trying again with .jpg

Here’s the “starting” image

Here’s the “final” image



I have an idea.
How about PIV(Optic flow)?



I think @hwada’s idea is really good as it can give you an idea of the velocities/the velocity field. But, it doesn’t necessarily segment each cell. It gives an overall representation. I haven’t used this, so you may want to give this a go. The issue is it won’t give you any indication of changes in morphology. More about PIV here.

I haven’t done this kind of analysis, but Ideas on other plugins or softwares you can use:
https://www.cellstar-algorithm.org/ (you will have to use cellprofiler for this)

Have you tried any of these??
Do let us know how you go.


Thank you for the continued interest and further insight Pradeep - appreciated. Sorry too for the lengthy delay to my reply - I didn’t realize there had been more activity around this post. I’ll certainly take a look at these other (possible) solutions to see if they can be adapted to my needs. I’ll strive to return here at some point when I’ve found a solution.

Best regards,