Hello all-- I’m new to ImageJ and trying to figure out how to count cells that are associated with specific antibodies I’ve used? I have the total cell count down but is there a way to only count cell associated/labeled with say green? Thanks in advance for your help.
Welcome to the Forum and ImageJ!
Let’s see. How did you go about counting your cells? Would you be able to post some images here to show us the datasets you are using for this analysis? With this info - we can see about helping you get a more specific solution for your datasets…
Though - here are some helpful links for getting started with ImageJ, Image Analysis, and Segmentation:
- ImageJ wiki - the best place to learn everything about ImageJ/Fiji!!
- “Introduction to Fiji” workshop - worth the time to get a solid intro
- Principles page - collection of principles for the entire image analysis process, from acquisition to processing to analysis
- Segmentation page
- “Segmentation in Fiji” workshop and corresponding slides
Eventually - you will need to write a bit of code to select only your ‘green’-labeled cells… but don’t be stressed about that if you’ve never written scripts before. ImageJ does try to make this relatively easy - here are some other helpful links for Scripting:
- Scripting overview page on the wiki
- A helpful workshop on Scripting with Fiji - the slides are here
- Introduction into Macro Programming page of the wiki
- Built-In Macro Functions list
I hope this helps! Post again when/if more questions arise too… we are all here to help!
I am not sure what image types you are using for your analysis, but one option that makes use of ImageJ indirectly is QuPath (https://github.com/qupath/qupath/releases/tag/v0.1.2). It is open source, and if your image type is supported, either initially or with the BioFormats plugin, it might be a quick way to solve your problem.
Just draw your region with one of the annotation tools, make sure your Image type is set to Fluorescence, and then run Analyze->Cell Analysis->Cell detection. You may want to play with the settings there, but the main one is determining which channel in your image is associated with DAPI. Then all you need to do is Classify->Classify by specific feature and read the results of your analysis through Measure-> Show annotation measurements. You will probably want to classify by something like “Mean cell channel X intensity.”
https://github.com/qupath/qupath/wiki/First-steps may also have some useful steps that I have skipped over here.
Hopefully it is acceptable to point out another useful open source solution! It seems to fit the bill here.
With the large number of views, I wanted to add some useful links for anyone pursuing the QuPath route:
Getting started from a user’s perspective:
Direct link to a multiplex classifier where you can set thresholds per channel and get results for any number of channels: