Hey everybody,
for my project I need to analyze the morphology of microglia, which I have previously stained (IHC) and scanned in a SlideScanner. Specifically, I need the ramification, number and length of branches and number and size of soma. I am using Fiji/ImageJ as a software, since our lab does not have a license for other softwares.
Thanks for your help :)
Hi,
Short answer- Yes you can definitely do it in Fiji
Longer answer:
I Don’t know how your Images look like or how many images you need to analyse so I will give you a little guide how I approached the analysis, but details might differ for your data.
Prepare your image for Analysis (depending on what’s needed, flat field correction, gauss filter smoothing, adequate stack reduction) you want to end up with a single image plane of homogeneous intensity so thresholding will work evenly over all of the image (especially necessary if you used a slide scanner and have huge images)
The whole trick is now to detect your microglia cells twice.
Detection one goes for the main body (your soma). Again, depending on your data blurring it, working with restrictive form factors and/or sizes might be all you need to get a nice body detection.
Run two goes for your entire cell, so all your branches need to be detected. This you might achieve with appropriate edge enhancement, stretching certain Intensity ranges and and/or again working with restrictive size and form factors.
Rest is than only working with ROIs and if you want generating an output that is right for you.
You have your body; whole cell minus body gives you the branches, which after skeletonizing and pruning will give you a readout of lengths, and branches.
It will only get tricky if you want the whole process to be automated because you have 100+ images, but still doable.
Good luck
Jürgen
Could you perhaps give more details on your images? Did you stain for Iba1 and nuclei? Is it fluorescence or DAB? How thick are your sections? Depending on the slice thickness, it might not be possible to reliably quantify the branches in an automated way.
*Edit: I'm asking because we're currently working on automated image analysis of microglia with Fiji. The script is written for 3 µm thick sections acquired at 10x with a Slide Scanner
Suckert Theresa
I stained for Iba1 only, non-flourescence. The sections are 30µm thick and acquired at 40x with the SlideScanner. Do you think it could still work?
- Badges
- Science method
- Similar topics
- Biological Science
- Histology
- Staining