Partial Integrated Density Measurement

[ademcan]

Dear List members,

I have a question about "integrated density" tool in 8 bit images.

Integrated density calculates total gray values by multiplying each gray
value with number of pixels with that value and adds them up. What I need is
to be able to determine a cut-off gray value and get the integrated density
of pixels above that value. This is very similar to thresholding in terms of
its mechanics. But instead of making everything 255 above the cut-off value,
I need to retain the gray value information and calculate integrated density
on those pixels above the threshold.

I cannot specify a ROI like in the dot blot method and get the integrated
density value for each ROI. I working on neuronal innervations, they don't
have regular shapes and spread all over the image. And also I am using batch
processing for image quantification, it is impossible to specify ROIs one by
one.

Here is a simple flow of how such function should work:

1. Open file from the specified directory
2. Determine the threshold value using certain predetermined parameters (e.g.
mean+2xstandard deviation) from the histogram
3. Calculate the partial integrated density of those pixels with values
above the threshold from Step 2.
4. Return partial integrated density value in the summary window with the
name of file
5. Go back to Step 1

I pretty much know how to get batch mode running in a set of images, only
thing missing is Step 3 and 4 for me. Thanks in advance for your help.


Adem Can

Ph.D. Student
Department of Psychology, Behavioral Neuroscience.
The University of Texas at Austin

[Christophe CHAMOT]

Adem Can a écrit :

> Dear List members,
>
> I have a question about "integrated density" tool in 8 bit images.
>
> Integrated density calculates total gray values by multiplying each gray
> value with number of pixels with that value and adds them up. What I need is
> to be able to determine a cut-off gray value and get the integrated density
> of pixels above that value. This is very similar to thresholding in terms of
> its mechanics. But instead of making everything 255 above the cut-off value,
> I need to retain the gray value information and calculate integrated density
> on those pixels above the threshold.
>
> I cannot specify a ROI like in the dot blot method and get the integrated
> density value for each ROI. I working on neuronal innervations, they don't
> have regular shapes and spread all over the image. And also I am using batch
> processing for image quantification, it is impossible to specify ROIs one by
> one.
>
> Here is a simple flow of how such function should work:
>
> 1. Open file from the specified directory
> 2. Determine the threshold value using certain predetermined parameters (e.g.
> mean+2xstandard deviation) from the histogram
> 3. Calculate the partial integrated density of those pixels with values
> above the threshold from Step 2.
> 4. Return partial integrated density value in the summary window with the
> name of file
> 5. Go back to Step 1
>
> I pretty much know how to get batch mode running in a set of images, only
> thing missing is Step 3 and 4 for me. Thanks in advance for your help.
>
>
>  

Hi Adem,

A simple macro can do it, if there are only images in your folder. Copy
and paste the following code in a macro (plugins>New...) and run :

-----------------------------------------snip------------------------------------------------------------
run("Set Measurements...", "  integrated limit redirect=None decimal=3");
dir1 = getDirectory("Choose Source Directory ");
list = getFileList(dir1);
setBatchMode(true);
for (i=0; i<list.length; i++) {
showProgress(i+1, list.length);
open(dir1+list[i]);
getStatistics(area, mean, min, max, std, histogram);
tmp=mean-(2*std);
if (tmp<0) {
min = 0;
}else{
min=tmp;
}
tmp=mean+(2*std);
if (tmp > 255) {
max = 255;
}else{
max = tmp;
}
//showMessage("Min : "+min+"\nMax : "+max);
setThreshold(min,max);
run("Measure");
}
-----------------------------------------snap------------------------------------------------------------

Cheers,

--
Christophe CHAMOT
---------------------------------------------------------------------
Plate-Forme de Recherche UMR7592
"Imageries des Processus Dynamiques
en Biologie  Cellulaire et Biologie du Développement "
Institut Jacques Monod, CNRS,  Universités Paris 6 et 7
2, place Jussieu - Tour 43
75251 Paris cedex  05
Tel: [phone]
fax: [phone]
http://www.ijm.fr/
---------------------------------------------------------------------