How to define areas of interest on an image and then quantify fluorescence in those areas of interest?

fluorescence
quantify
dapi

#1

Hi all,

I am an undergrad just getting started with ImageJ and I am unsure of how to quantify cell fluorescence on some images I have taken.

For context: I stained cells with two fluorophores, DAPI and PerCP. I currently have .lif image files of these slides, where each image file contains three different, adjacent images: one taken with visible light, one at DAPI emission wavelength, and one at PerCP emission wavelength. I have installed the BioFormat plugin for ImageJ so that the software is able to open the .lif image files.

I would like to use the DAPI image to define areas of interest on the slide (i.e. I am only interested in analyzing the PerCP fluorescence of cells that displayed DAPI staining). Once I have used the DAPI image to select areas of interest, I would like to quantify the level of PerCP staining in those areas of interest, perhaps by converting the PerCP image to greyscale and measuring the brightness.

I have to admit I really have no idea how to start. I’ve googled around a bit but have not had success at accomplishing this task, I think I just lack the software familiarity needed to engage in trail-and-error intelligently, so I’ve just been floundering around not knowing what’s going on.

Does anyone have any tips on how to:

  • select only regions that exhibited DAPI fluorescence for further fluorescence quantification analysis?
  • convert PerCP fluorescence to greyscale and measure the brightness in the areas of interest?

Thanks so much!


#2

Godd day!

I think what you like to do has been done already several times by others.

Did you search this Forum by terms like

flourescence, DAPI, double staining, colocation, etc.

Furthermore and instead of asking Google, it is always a good idea to focus on the tool in question and first study the ImageJ User Guide:
https://imagej.nih.gov/ij/docs/guide/index.html

After having done this we are happy to answer questions beyond the basics!

Regards

Herbie


#3

Hi, thanks for pointing me to the ImageJ User Guide! Following the advice from that document, I’ve been able to select multiple regions of interest simultaneously (using the ROI Manager) and I’ve been able to convert one region of interest to grayscale at a time.

I do have two follow-up questions though:

  1. I selected oval ROIs, but when I convert an ROI to grayscale (via Image --> Transform --> Image to Results) the Results table lists grayscale values for a rectangular grid of numbers. I believe the rectangular grid of numbers is such that the oval region of interest is inscribed within the rectangular grid listed in the Results table. Is there any way I can restrict the Results table to only provide the grayscale values of pixels within the oval region of interest I’ve selected?

  2. I am able to create multiple regions of interest at a time using the ROI Manager, but I have only been able to convert one ROI at a time to grayscale values. Is there any way to do ROI --> grayscale for multiple distinct ROIs in an image simultaneously?

Thanks!


#4

There is a common question regarding both of your points that needs to be answered first:
Why do you want to get an array of numbers from the ROIs?

Did you try splitting your image intoo teh RGB channels?
Did you try the “CMYK”-plugin.
Did you consider the “Colour Deconvolution”-plugin?

Try to find out about those ways.

ROI to grayscale
doesn’t require to do what you presently do.
Please study the User Guide which will take hours to days but will really help with your problems.

Regards

Herbie