Help with total area tissue measurement

[sdegan]

Hi All,

I am having trouble calculating the total tissue area in
pathological slides. What I do is: I convert the image to 'Image type
16-bit' then I run the automatic threshold to cover the total area of
the tissue. The problem is that the threshold system does not overlap
with the total tissue area (i.e. tissue area>threshold). If I
manually adjust the threshold I see an increase of the  background on a
side of the image. To note: there is no background on the images. If I
also convert the image to 'Binary'  and 'grow' I still do not have
certain area of the tissue covered (perhaps because these area are not
as dark as the others?).

On the same slides I have successfully calculated the cells
that are stained by an antibody using the Color deconvolution method
but now I need to carry out the percentage calculation of the cells
stained over the total area of the tissue.

Can someone help me with this?

Many thanks

simone

[Joel Sheffield]

Hi Simone

In the absence of an image to look at, I would think that you have an
uneven background, either because of illumination or sample
irregularities.  I have found that the bandpass filter, set at
"filter out pixels greater than 100, and less than 0" often smoothes
out the background so that you can make the kind of measurement you
want.  Uncheck all of the contrast adjustment options.

Best of luck,

Joel

> Hi All,
>
> I am having trouble calculating the total tissue area in
> pathological slides. What I do is: I convert the image to 'Image type
> 16-bit' then I run the automatic threshold to cover the total area of
> the tissue. The problem is that the threshold system does not overlap
> with the total tissue area (i.e. tissue area>threshold). If I
> manually adjust the threshold I see an increase of the  background on a
> side of the image. To note: there is no background on the images. If I
> also convert the image to 'Binary'  and 'grow' I still do not have
> certain area of the tissue covered (perhaps because these area are not
> as dark as the others?).
>
> On the same slides I have successfully calculated the cells
> that are stained by an antibody using the Color deconvolution method
> but now I need to carry out the percentage calculation of the cells
> stained over the total area of the tissue.
>
> Can someone help me with this?
>
> Many thanks
>
> simone

--
Joel B. Sheffield, Ph.D.
Biology Department, Temple University
1900 North 12th Street
Philadelphia, PA 19122
[hidden email]  
[phone], fax [phone]
http://astro.temple.edu/~jbs

[karo03]

If the image consists of tissue only, no background, then calculate  
the image area by "Select All" followed by "Measure". If there is  
background then segmentation of either background or tissue should be  
possible, in worst case by hand. In the first case tissue area is  
whole image area minus background area. By hand you should select  
always the most easy part.
Regards
Karsten

Am 27.01.2009 um 17:03 schrieb Simone Degan:

> Hi All,
>
> I am having trouble calculating the total tissue area in
> pathological slides. What I do is: I convert the image to 'Image type
> 16-bit' then I run the automatic threshold to cover the total area of
> the tissue. The problem is that the threshold system does not overlap
> with the total tissue area (i.e. tissue area>threshold). If I
> manually adjust the threshold I see an increase of the  background  
> on a
> side of the image. To note: there is no background on the images. If I
> also convert the image to 'Binary'  and 'grow' I still do not have
> certain area of the tissue covered (perhaps because these area are not
> as dark as the others?).
>
> On the same slides I have successfully calculated the cells
> that are stained by an antibody using the Color deconvolution method
> but now I need to carry out the percentage calculation of the cells
> stained over the total area of the tissue.
>
> Can someone help me with this?
>
> Many thanks
>
> simone

Karsten
[hidden email]