Z Depth / Resolution

simpleneuritetracer
Tags: #<Tag:0x00007fd542818958>

#1

Hello,

I would like to correctly identify the x, y and z resolution of old 2-photon image stacks. How do I retrieve this information from the metadata to calibrate my images? Is the z resolution the same as ImageJ’s voxel depth? I am still fairly new to microscopy. Thank you for your help!


#2

Are your images in dicom format? If so, you could lookup for this information at it.

In my images of mammography the information on Fiji’s menu Image > Properties… are the same as in dicom tag: 0028,0030 which is related to “Pixel Spacing”.


#3

Hi @IR23 isabel.

The optical resolution of the 2 photon microscope is determined by the illumination wavelength and the numerical aperture of the objective lens

The image resolution depends on how the scan we set up.

ImageJ will try to read the image meta data and after opening the image will.tell you in image properties window what it read for the xy and z image resolutions.

In theory the optical resolution should be oversampled about 2X by the image resolution. But it’s some work to find out if that was the case.

If you tell us what imageJ reads the xy and z Pixel sizes (really it’s spacing not Size)
And the illumination wavelength and objective NA, we can calculate the optical resolution and see if it’s sampled nicely by the image resolution.

Read handbook of confocal microscopy by James Pawley for more details.

Cheers

Dan


#4

Hi Chalkie,

Thank you for the information! This is displayed in the ImageJ properties window:
X & Y = 1.1655 microns
Z = 1 micron

It is believed these images were taken with a Z of 5 um, but I am trying to verify if it’s correct. This neuron was previously traced in Neurolucida. I traced the same neuron using the Simple Neurite Tracer in ImageJ and there is an increase of 4,000 um lengthwise. I am not sure if the algorithms are different or if the XYZ are incorrect. Please see the images of the meta data below. The NA is 0.95. Thank you very much!


#5

Hi Helder,

Thank you for your reply!

I attached some images below your post. Than you for your help!


#6

Hi @IR23, it seems your metadata (xml) file doesnt contain any information about pixelsize in Z…

I’m sorry to not be more useful…


#7

@IR23 the last Screenshot of the XML Metadata shows the z step between images was 5 microns. Look at the X y z positions of consecutive images in there.

So the meta data is saying what you expect but imageJ is not getting that correct z step metadata into the image properties correctly. It’s still at the default of 1.

What to do?
It might be a bug. So you can report that using the imageJ bug reporter in help menu. You can paste in a link to this topic on the forum for reference.

For now, you can simply reset the image properties z spacihg value correctly manually. Write 5 in there then clock ok and then re do the analysis. Distances In z will now be correct :wink:


#8

Thank you very much!