Automatic Morphology Analysis for a Set of Images in ImageJ (batch processing)

imagej
batch-processing
morphology
cellsize

#1

Hi all,

I have a big set of images from high-throughput microscopy that I need to analyze the morphology of the cells (cell size, roundness, aspect ratio, etc.). Is there a way to do that for all images simultaneously in imageJ?

Note that images are taken with bright-field.

I would welcome suggestions of other softwares if this is not possible with imageJ.

Thank you!


#2

Good day!

Is there a way to do that for all images simultaneously in imageJ?

If you have as many CPU-cores as you have images, perhaps.

What you mean is automatically perform batch prcessing, one image after another but without manual intervention — no?

This is in principle possible with ImageJ/Fiji.

Without seeing any typical (not the best) images in PNG- or TIF-format* it is impossible to tell you if it is actually possible!

Regards

Herbie


* No JPG-format though, because JPG introduces artifacts!
(Converting a JPG-compressed image to TIFF- or PNG-format doesn’t make sense.)
You may also post images as Zip-archives or make them accessible via a dropbox-like service.


#3

This is an example:
https://drive.google.com/file/d/1JVjO4t9gMqP8tnPmP6ARbcPoGZQO3h-U/view?usp=sharing

I have a huge amount of those images.
Could you please explain how to do batch analysis in imageJ for such images. Would be great if I can get the results as an excel sheet.


#4

The illumination of your sample image is clearly uneven. The microscope isn’t set up correctly. Please consult a professional microscopist!

Regards

Herbie


#5

Thank you for the advise. I will check this with the microscopist.

Would you be able to provide a general procedure on how to measure cell morphological parameters for a number of images at once (bright-field images)? I know how to get that for one image at a time, but not for a batch of images. Any specific plug-in to perform the desired task given that I managed to get more professional images?


#6

Hi,

I think that Cellprofiler is a good alternative to ImageJ for your pictures. It processes pictures batchwise. There is also a module to reduce uneven illumination.

Regards,
Peter


#7

Would you be able to provide a general procedure on how to measure cell morphological parameters for a number of images at once (bright-field images)?

This demo-macro tells you how to do that.

Here is what I get from your sample image:

Please note that I had to slightly crop your image due to the illumination problem.
865 cells have been detected and here
Results.zip (22.6 KB)
is the list as a zipped text file that contains the descriptors of these cells.

Please carefiully study the ImageJ User Guide to get the meaning of the descriptors.

An ImageJ-macro that does all the processing isn’t difficult to write. For a first try use the macro recoder functionality of ImageJ.

Reduction of uneven illumination is always a sub-optimum approach. I used it for the above results but you really should improve image acquisition because you will get better and more robust results in the end.

Regards

Herbie


#8

the Results file you provided is exactly what I’m looking for.

When running your macro (plug-ins --> macro --> record --> add ur macro and the directory in which my files are saved), I do not get the information as in your results file. All I get is 4 columns (Area, Mean, Min, and Max).

Excuse my limited imageJ knowledge as I’m still a beginner.


#9

The demo macro is not mine and it should only show you how batch processing is to be done with an ImageJ macro. Of course you have to enter your specific processing code to it!

Because you wrote that you know how to process a single image I thought that this information will help you to obtain batch processing.

Please read carefully what I’ve written:

An ImageJ-macro that does all the processing isn’t difficult to write. For a first try use the macro recoder functionality of ImageJ.

Perhaps you have to learn how to write ImageJ-macros.
https://imagej.nih.gov/ij/developer/macro/macros.html
https://imagej.nih.gov/ij/developer/macro/functions.html

Presently and because you don’t have better images, I think it makes little sense to write a macro. Just tell us when you have better images and we will try to help with coding.

Regards

Herbie