3D nuclei quantification

Dear forum members, can someone suggest an optimal tool for quantitative volumetric analysis of nuclei shape&structure for hdf5 image? This is an Arabidopsis lateral root staining with DAPI and scanned with 3D.
I am attaching an example.
I appreciate any help you can provide. ### Sample image and/or code

Good morning Taras,

If I have correctly understood your problem, I suggest you to use the ImageJ pluggin : “3D Image Suite”. (Here is a link for an introductory video : https://www.youtube.com/watch?v=OPC2kP-5By4).

The idea would be to threshold the image (to catch all the nuclei and fill them if possible), then use the 3D Simple Segmentation function to obtain the volume (number of voxels that you will multiply by the voxel dimension). This is a link to a (not so far ?) problem I have encoutered and resolved : Ventricular Cavities 3D Volumes Measurement on ImageJ

Hoping it will help you,
Nabil Nicolas.

NB : Also, before beginning the image treatment, you can apply a (3D) median filter to smooth a little bit the structures.

Thank you very much!

1 Like

Thanks again! I tried segmentation with HRPPP, it works quite well, but since the image is 3D with up to 250 µm thickness, there is a problem with an equal threshold. But if I will have 50% of the organs OK, it will be enough. I can extract all parameters from HRPPP, volume, shape etc. really nice!

1 Like

You’re welcome !

Concerning the rest of the problem, there is another software (paying) : Imaris that might help you thanks to fonction called “Surface”. This function alows you to threshold, cut connections between objects, and add filters to eliminated some non-desired objects on 2D, 3D and 4D images. And at the end of the treatment, you will obtain the (specific & average) volume of your object(s), the area, …

Dear Nicolas, thanks! I know Imaris, but HRPP is open source and have more possibilities as Imaris in the term of volumetric analysis, segmentation, particle recongnition, 3D rendering etc. You can look on example of application: https://www.embopress.org/doi/full/10.15252/embj.2020104416
My best regards!

1 Like