measure shortest distance

[Jelle van den Ameele]

 
Dear all,
 
I would like to measure how far every single cell of a group of cells is
lying from the blood vessels around. So I need the shortest distance between
the center of  a  particular cell and the nearest border of a blood vessel.
I have the cells of interest in green and the blood vessels in red, so I can
split the color-channels and generate tresholded images if necessary, but
don't get any further.
Unfortunately I don't have any programming skills and am rather new in
ImageJ, so I was wondering whether someone could help me a bit further?
 
Thanks in advance!
 
Jelle van den Ameele
Aspirant FNRS
Institute for Interdisciplinary Research (IRIBHM)
Université Libre de Bruxelles (ULB)
Campus Erasme
Route de Lennik 808, C6.111
1070 Bruxelles, Belgium
Lab: [phone]
Mobile: [phone]
E-mail: [hidden email]

[Kyle E. Miller]

Dear Jelle,



One approach is to take the blood vessel image, threshold it (invert the lut
if needed to make the blood vessels black) and then make a distance map
using the command process > binary > distance map. The pixel intensity now
gives the shortest distance from the blood vessels. Take the image of the
cells and then use the process > find maxima > output single points. You
might need to adjust the noise, for example if you use a thresholded image
input 0.

This returns individual pixels where each of the cells are positioned. Then
use process > math > divide and enter 255. The background should be equal to
0 and the individual pixels will have an intensity of 1.


Then do  process > image calculator and multiply the two images. You now
have a new image where the pixel intensity of each of the pixels
representing the cells is equal to the shortest distance to the edge of a
blood vessel.



Follow this with analyze > histogram. Then hit list. This gives a table that
can be copied to excel and gives the shortest distance of each cell to a
blood vessel.


Best wishes,
Kyle

On Thu, Nov 18, 2010 at 4:15 PM, Jelle van den Ameele <
[hidden email]> wrote:

>
> Dear all,
>
> I would like to measure how far every single cell of a group of cells is
> lying from the blood vessels around. So I need the shortest distance
> between
> the center of  a  particular cell and the nearest border of a blood vessel.
> I have the cells of interest in green and the blood vessels in red, so I
> can
> split the color-channels and generate tresholded images if necessary, but
> don't get any further.
> Unfortunately I don't have any programming skills and am rather new in
> ImageJ, so I was wondering whether someone could help me a bit further?
>
> Thanks in advance!
>
> Jelle van den Ameele
> Aspirant FNRS
> Institute for Interdisciplinary Research (IRIBHM)
> Université Libre de Bruxelles (ULB)
> Campus Erasme
> Route de Lennik 808, C6.111
> 1070 Bruxelles, Belgium
> Lab: [phone]
> Mobile: [phone]
> E-mail: [hidden email]
>

--
Kyle E. Miller
Assistant Professor
Department of Zoology
Neuroscience Program
Cell and Molecular Biology Program
Center for Mitochondrial Science and Medicine

Michigan State University
336 Natural Sciences Bldg.
E. Lansing, MI 48824

Office Phone: [phone]
Lab Phone:    [phone]
Cell Phone:   [phone]
Lab website: http://millerlab.zoology.msu.edu/

[mercedes1225]

Hi Kyle,

I'm doing my final degree project in ImageJ and am totally new to it. I'm trying to achieve a similar goal to Jelle: I have stacks of RGB images of osteoblasts (in the green channel) and stem cells (in the red channel). I'm trying to find the shortest distance of a stem cell from its nearest osteoblast. I've tried your method and have also tried altering it to make it a little more appropriate, as I want to measure distance form the edge of the stem cell to the nearest edge of an osteoblast. Any chance you can help out?

Cheers,
Grace