3D Region of Interest

[BirthingXS]

Hi all,

I'm trying to do some 3D clustering analysis described below, and was wondering if anyone has ideas how it can be done:

1. Counting 3D clusters in an image using '3D Object Counter'.
2. Import the results into 3D Roi Manager, selecting all the objects which places a Roi around each cluster.
3. The Roi is then enlarged by a set number of pixels to incorporate some of the neighboring clusters as a single cluster.
4. The question is: Is there a way to see which of the original clusters have been included in to the new clusters?

Edit: Another problem is the calculation of NND in 3D. The Distances function in 3D Roi manager computes the distances between selected pairs of objects. However, with a sizable number of objects, this simply takes too long. Is there an option to measure only the closest neighbors instead of all the distances?

Thanks in advance.

XS

[Thomas Boudier-4]

Hi XS,

Are you referring to the Ripley's K function ?

http://icy.bioimageanalysis.org/plugin/Spatial_Analysis

Best,

Thomas



On 29/11/2016 18:45, BirthingXS wrote:

> Hi all,
>
> I'm trying to do some 3D clustering analysis described below, and was
> wondering if anyone has ideas how it can be done:
>
> 1. Counting 3D clusters in an image using '3D Object Counter'.
> 2. Import the results into 3D Roi Manager, selecting all the objects which
> places a Roi around each cluster.
> 3. The Roi is then enlarged by a set number of pixels to incorporate some of
> the neighboring clusters as a single cluster.
> 4. The question is: Is there a way to see which of the original clusters
> have been included in to the new clusters?
>
> Thanks in advance.
>
> XS
>
>
>
> --
> View this message in context: http://imagej.1557.x6.nabble.com/3D-Region-of-Interest-tp5017648.html
> Sent from the ImageJ mailing list archive at Nabble.com.
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

... [show rest of quote]

--
   /***************************************************************/
      Thomas Boudier, Associate Professor, UPMC,
      Université Pierre et Marie Curie, Paris, France.
      BII (BioInformatics Institute)/UMI 2955 IPAL, Singapore.
/**************************************************************/

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[BirthingXS]

Hi Thomas,

I'm having some trouble with loading some of the contents of the link, are you able to elaborate on what the function does (Sorry as I'm not a programmer myself). Also if it helps, this is the function I'm using for 2D and would like it to be translated to 3D.

setOption("BlackBackground");
original = getTitle();
run("Invert");

roiManager("reset");

run("Create Selection");
run("Enlarge...", "enlarge=2");
roiManager("Add");
roiManager("Split");

roiManager("select", 0);
roiManager("Delete");

selectWindow(original);

run("Invert");
run("Set Measurements...", "area stack display redirect=None decimal=3");
for(i=0; i<roiManager("count"); i++) {
        roiManager("Select", i);
        roiManager("Measure");
        //if the selection is empty (included holes) the measurement and result will be deleted
        if(getResult("%Area", nResults-1)==0) {
                roiManager("Delete");
                IJ.deleteRows(nResults-1, nResults-1);
                i--;
        //if there is a particle inside the selection the CRU and tne clusters will be analyzed
        } else {
                setResult("Label", nResults-1, getResultLabel(nResults-1) + "-CRU");
                run("Analyze Particles...", "display");
        }
}

Additionally, I've been in touch with Jan Brocher with some of these issues and he suggested you might be able to offer some insights. Do you mind if I get in touch with you via email?

Cheers,
Xin

[Thomas Boudier-4]

Hi Xin,

The macro will basically cluster together closest objects. If A is
closest to B and B is closest to C, the algo will cluster A B and C.
This is a very simple algorithm but can be eventually useful to detect
clusters.

Can you send me directly some sample image to test this macro.

Best,

Thomas


On 29/11/2016 19:47, BirthingXS wrote:

> Hi Thomas,
>
> I'm having some trouble with loading some of the contents of the link, are
> you able to elaborate on what the function does (Sorry as I'm not a
> programmer myself). Also if it helps, this is the function I'm using for 2D
> and would like it to be translated to 3D.
>
> setOption("BlackBackground");
> original = getTitle();
> run("Invert");
>
> roiManager("reset");
>
> run("Create Selection");
> run("Enlarge...", "enlarge=2");
> roiManager("Add");
> roiManager("Split");
>
> roiManager("select", 0);
> roiManager("Delete");
>
> selectWindow(original);
>
> run("Invert");
> run("Set Measurements...", "area stack display redirect=None decimal=3");
> for(i=0; i<roiManager("count"); i++) {
> roiManager("Select", i);
> roiManager("Measure");
> //if the selection is empty (included holes) the measurement and result
> will be deleted
> if(getResult("%Area", nResults-1)==0) {
> roiManager("Delete");
> IJ.deleteRows(nResults-1, nResults-1);
> i--;
> //if there is a particle inside the selection the CRU and tne clusters will
> be analyzed
> } else {
> setResult("Label", nResults-1, getResultLabel(nResults-1) + "-CRU");
> run("Analyze Particles...", "display");
> }
> }
>
> Additionally, I've been in touch with Jan Brocher with some of these issues
> and he suggested you might be able to offer some insights. Do you mind if I
> get in touch with you via email?
>
> Cheers,
> Xin
>
>
>
> --
> View this message in context: http://imagej.1557.x6.nabble.com/3D-Region-of-Interest-tp5017648p5017650.html
> Sent from the ImageJ mailing list archive at Nabble.com.
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

... [show rest of quote]

--
  /***************************************************************/
       Thomas Boudier, Associate Professor, UPMC,
       Université Pierre et Marie Curie, Paris, France.
       BioInformatics Institute (BII)/IPAL, Singapore.
/**************************************************************/

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html