Using 3D object counter data to find slice and export

Image Analysis
, ,
#1

Hello everyone!

First off! Thank you so much for the help in the past at answering my questions and queries. You’ve really helped me and given this lowly undergraduate a lot to do and enough hopefully to get a summer internship in my lab.

My question today is an idea I’ve had for building a classification machine learning system, I just need some help getting images from imageJ as output data.

Having used the 3D object counter and the 3D object Manager - Is there a method I can use that will:

  1. for each object find it’s centre
  2. Locate the slide it is in
  3. Jump to that slice
  4. Export slice as a flat image with a specific name i.e. “object4.png”

Many thanks if anyone is able to help! Will make my life so much easier when I hit process for my 1000 or so images.

Nahtan.

#2

Dear @SirSharpest,

That’s a matter of definition, since the centroid could lie in between two slices. You’d have to extract the coordinate in Z dimension and round to the closest slice in general. The 3D Manager, however, seems to extract the centroid in such a way, that it is located on a slice.

You can run the following macro on an image that you have segmented via 3D Manager’s 3D Segmentation functionality:

id = getImageID();
run("3D Manager"); // Enable extensions
Ext.Manager3D_AddImage(); // Add segmented image to 3D Manager
Ext.Manager3D_Count(nb); // Count objects for iterating
for(i=1;i<nb;i++) {
  Ext.Manager3D_Select(i);
  Ext.Manager3D_Centroid3D(i,cx,cy,cz); // Compute centroid 
  selectImage(id); // Get input image
  Stack.setSlice(cz); // Set slice accordingly -- might be problematic if the input image is a hyper stack
  run("Duplicate...", "use"); // Duplicate from stack
  Ext.Manager3D_GetName(i, name); // Get object name from 3D Manager
  saveAs("/Users/stefan/Desktop/test/"+name+".tif"); // Save to folder -- redefined folder
  close();
}

If you need additional information, feel free to post again :wink:

Best,
Stefan

2 Likes
#3

Hey!

Thanks so much for the reply. Can’t wait to give this a try when I’m back in the office.
Will let you know, seems to be exactly what I want :slight_smile:

Thank you so much,
Nathan