How to measure volume and suface of a cell filmed in a .tiff 4-d stack

[Daniele Vilone]

Hi,

I am doing a research about cell shape modifications. I have a 4-dimensional stack (each time frame is made by 34 2-d images each one representing a slice along the z axis) of a moving cell (the "movie" is a .tiff file). I should measure the volume of the cell and the surface of its external membrane. How can I do with icy? I'm trying with the Active Cells 3D plugin, but I can't find the way to do that. I apologize for this probably trivial question, but I am very new with icy (and also in a hurry...).

Thank you very much

Daniele

[Alexandre Dufour]

Hi Daniele,

Here is a protocol that will let you do this (you may open the attached .xml file via the Protocols plugin of Icy).

0) First make sure install the "Track Processor mesh analysis" plugin (via Icy's search bar or Icy's plugin browser)

Basically what this protocol does when you click on "run" is:

1) find a rough estimate of the cell(s) in the first frame (via the HK-Means block). Here make sure the min/max volume of your cells (in voxels) is relevant to your data (my default values may be incorrect for you)

2) construct an active contour in the form of a triangular mesh (hence the name "3D Active Mesh").

3) Adapt the mesh such that it fits cell surface. Here there are a number of parameters to adjust, but I am in the process of simplifying the interface in future versions. In the mean time, the protocol gives you default parameters that work in many cases. If you have trouble setting them up, let me know and we'll work them out together.

4) repeat this fitting process for each time point (automatic)

5) Once tracking is done, the "Track Manager" plugin will open, and show you a timeline of the tracking results. In this window, click on "Add processor" (this list contains various tools to measure tracking data) and select "Track Processor Mesh analysis" This will let you plot various parameters including volume/surface of tracked objects over time and export this to Excel.

And if you want to test this protocol to see it in action, you may try it on the attached data set (4D tiff file).

Let me know if this helps,

Alexandre

[louise-laure]

Hi Alexandre,
I just tried out your protocole on your image and it reports a bug (I clicked on the report bug button)
Hope it's not a big deal,

LL