Quantifying difference in fish skin histology

Hello. I am very new to Fiji and image analysis. I want to find out what my options are for quantifying differences in cell types and thicknesses of fish epidermis and dermis in fish that are being used for a wound-healing experiment using cleaner shrimps. Fish are being processed for standard H&E histology to identify normal tissue, initial wound initiation (removal of scales from the epidermis), and 2, 4 and 7 days post wound initiation to detect healing.

To begin with I would like to use Fiji in a pilot study to determine the variation in a sample of 10 individual fish to confirm that my sample size is good. Thereafter, I will begin the experiment.

Can anyone offer me any advice please on a simple method for quantifying the different cell types and thicknesses of regions in a standard H&E section of fish skin? Are there any protocols that have similarly been written for this that I could use and practice with to start? My knowledge of image analysis is currently very basic.

Any help is greatly appreciated.

Yours sincerely, David

Hi @dvaughan

I have recently worked with H&E stained images, and had to realize
that there is no one solution fits all for these kind of images. The
segmentation method applied prior to quantification heavily depends on
your images.

I tried multiple approaches;

  1. Using k-Means on the images to identify the different structures

  2. Doing a naive segmentation (after some color adjustment) then using a predictor to separate good from bad labels.

But the result really depends on the images (H&E tends to be non uniform)

If you have an example image(s), it would be easier to help you.


Hi Oole

Thank you for the response.

I will be doing a pilot study in the next week which will run for 7 days before I can get image examples to you. Please let me know if this is ok with you?

Best regards


Hi @dvaughan

Of course!