I'm doing an invadopodia assay, and from what I found on the literature, people use x63 magnification of the confocal microscope to find a specific area, image it and use ImageJ to analyse the degradation area over the area covered by cells. But one thing that I found is that x63 when image only cover a very small fraction of the observing field when the software snaps the image. So this makes the quantification quite biased and nonrepresentative in my opinion.
Do you think I can use the smaller magnification (like x40) to take the image to analyse? I asked some people and they said x smaller than 63 doesn't give enough resolution so the quantification will not be accurate. But I would argue that this applied to the whole image so there shouldnt be any change? And also can ImageJ still distinguish the degradation area from the surrounding with lower magnification than 63?
Thank you very much
Kenth: Oh yeah, you're right. I completely forgot about the difference between resolution and magnification. Thanks for reminding me that.
I'm not sure about the digital zoom though. I've just started to use the Zeiss 710 confocal recently so I don't really know much about the software as well as all the functions it has yet. But I can ask the scientific officer about what you suggested. Thank you very much.
Concerning ImageJ analysis, yes for sure you can analyse images taken with smaller magnification. The only concern would be that if your microscope doesn't have a motorized revolver (meaning that you change manually the objective), make sure to calibrate correctly the size of your pixel depending on the magification and zoom. Also consider that if you want to compare apple with apple, all your images to compare should be taken with the same parameters (same objective, zoom, PMT voltage, gain, offset, laser power, etc.)
To calibrate your pixel in imageJ, go to Image > Properties; or to Analyse > Set scale.
Good luck!
Monique
Monique: Thank you for the answer. But may i ask further why do you need to calibrate your photos? If you know that all your photos have the exact same number of pixels, will you still need to calibrate them? In other words, when do you need to calibrate? And what calibration actually does to your image?
Thank you very much
You can take 2 differents images of, for example, 514x514 pixels. This means that both images have the same number of pixels but this doesn't mean that they have the same pixel size (µm/pixel). For instance if you took one image 514x514 at zoom 1 and the other 514x514 at zoom 3, In size (µm), one pixel will be 3x smaller than the other one.
Yes, this is perfectly true for most microscope. The exception comes with old non-motorized microscope where the user has to change himself the objective :)
Monique: Ok, but I don't think i change anything during the image-taking process to be honest. I don't see the point of changing the zoom to make it more complicated for my analysis at the moment. So I dont think I need to calibrate.
Kenth: i just checked with the Bio-Formats plugin. Usually I imported the image to ImageJ by dragging the file .lsm to the software and then using the command "Stack to Images" to separate the image from different channels. All images at different locations and different channels were taken at the exact same set up with the other. I checked with the Bio-format plugin and it gave me the same photos as when I use "Stack to Images" command, all said 134.95x134.95 um (1024x1024), 16-bit, 2MB. So I assume the two command did the same thing or I dont really need to calibrate?
OK no, you don't need to calibrate. Your image is calibrated since you have µm mentioned in your image header.
I see. That sounds just great to me! Thank you two very much. I've just started my PhD and quite new to image analysis and integrity so I just need to make sure I get to the core and fundamental.
You may have a look to the attached link if you are interested in image analysis concepts and guidelines using ImageJ.
https://www.researchgate.net/post/co-localization-Z-stacked_images_Leica#5853bfbcf7b67ebf202bd751
- Badges
- Science topic
- Similar topics
- Engineering
- Materials Engineering
- Powders