GUS staining quantification

[magi78343]

Dear all,

I need to quantify the GUS-staining intensity (blue) of some digital images. As far as I know I first need to split the RGB channels. I am interested only in the blue intensity but the blue channel is the pilest one and the dark gray color corresponding to the dark blue on the color image remains in the red channel. Which channel should I use for measurement?

Thank you in advance.

Magi

[Stoyan Pavlov]

Hello Magi,
Can you attach a sample image?
When I was starting with image acquisition and analysis I had similar
problem (i guess): My white balance was off somehow and the acquired images
of DyI stained tissue were oversaturated in the red channel, however in the
green the dynamic range was just fine, so I used the green channel instead.
As long as there are no multiple labels visible while you acquire the image
you can easily use another channel (still all your images must be acquired
the same way and you'll need to use the same channel in order to be able to
compare them).
Stoyan


2013/3/25 magi78343 <[hidden email]>

> Dear all,
>
> I need to quantify the GUS-staining intensity (blue) of some digital
> images.
> As far as I know I first need to split the RGB channels. I am interested
> only in the blue intensity but the blue channel is the pilest one and the
> dark gray color corresponding to the dark blue on the color image remains
> in
> the red channel. Which channel should I use for measurement?
>
> Thank you in advance.
>
> Magi
>
>
>
> --
> View this message in context:
> http://imagej.1557.n6.nabble.com/GUS-staining-quantification-tp5002411.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]

--
Dr. Stoyan P. Pavlov, MD, PhD
Departament of Anatomy, Histology and Embryology
Medical University "Prof. Dr. Paraskev Stoyanov", Varna
Prof. Marin Drinov Str.55
9002 Varna
 Bulgaria
Tel: +359 (0) 52 - 677 - 086
e-mail: [hidden email]

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

[magi78343]

Hi Sergey,
Here is a nice example of the images I'm working with.  I have to measure the mean intensity of the whole blue area in the tip by drawing ROI around it (my supervisor doesn't like the tresholding)



Best,
Magi

[G. Esteban Fernandez]

Hi Magi,

I would convert from RGB to HSB (Hue, Saturation, Brightness) and measure
the intensity of the Saturation channel.

Generally, Saturation reflects the amount of dye present.  Brightness
depends more by the density of the tissue and Hue = Color (since you have
one consistent dye/color the Hue doesn't change).

-Esteban


On Tue, Mar 26, 2013 at 11:49 AM, magi78343 <[hidden email]> wrote:

> Hi Sergey,
> Here is a nice example of the images I'm working with.  I have to measure
> the mean intensity of the whole blue area in the tip by drawing ROI around
> it (my supervisor doesn't like the tresholding)
>
> <
> http://imagej.1557.n6.nabble.com/file/n5002445/AUX1-GUS_1%C2%B5M_ACC_8.jpg
> >
>
> Best,
> Magi
>
>
>
>
>
> --
> View this message in context:
> http://imagej.1557.n6.nabble.com/GUS-staining-quantification-tp5002411p5002445.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]

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

[G. Esteban Fernandez]

You can convert to HSB with:  Image > Type > HSB Stack.

-Esteban


On Tue, Mar 26, 2013 at 12:54 PM, G. Esteban Fernandez <
[hidden email]> wrote:

> Hi Magi,
>
> I would convert from RGB to HSB (Hue, Saturation, Brightness) and measure
> the intensity of the Saturation channel.
>
> Generally, Saturation reflects the amount of dye present.  Brightness
> depends more by the density of the tissue and Hue = Color (since you have
> one consistent dye/color the Hue doesn't change).
>
> -Esteban
>
>
> On Tue, Mar 26, 2013 at 11:49 AM, magi78343 <[hidden email]> wrote:
>
>> Hi Sergey,
>> Here is a nice example of the images I'm working with.  I have to measure
>> the mean intensity of the whole blue area in the tip by drawing ROI around
>> it (my supervisor doesn't like the tresholding)
>>
>> <
>> http://imagej.1557.n6.nabble.com/file/n5002445/AUX1-GUS_1%C2%B5M_ACC_8.jpg
>> >
>>
>> Best,
>> Magi
>>
>>
>>
>>
>>
>> --
>> View this message in context:
>> http://imagej.1557.n6.nabble.com/GUS-staining-quantification-tp5002411p5002445.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]

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

[Sergey Gutor]

Hi Magi,

If you selected area without thresholding you can do it on native image or
on any channel in RGB or HSB stack. It is better on Red channel in RGB and
Saturation in HSB for my eyes. In you picture blue is not "blue" for your
computer. Clean Red/Green/Blue colors for you PC look on the picture in
attach. Open this picture in ImageJ and look it in RGB stack.

Best regards,
Sergey


2013/3/27 G. Esteban Fernandez <[hidden email]>

> You can convert to HSB with:  Image > Type > HSB Stack.
>
> -Esteban
>
>
> On Tue, Mar 26, 2013 at 12:54 PM, G. Esteban Fernandez <
> [hidden email]> wrote:
>
> > Hi Magi,
> >
> > I would convert from RGB to HSB (Hue, Saturation, Brightness) and measure
> > the intensity of the Saturation channel.
> >
> > Generally, Saturation reflects the amount of dye present.  Brightness
> > depends more by the density of the tissue and Hue = Color (since you have
> > one consistent dye/color the Hue doesn't change).
> >
> > -Esteban
> >
> >
> > On Tue, Mar 26, 2013 at 11:49 AM, magi78343 <[hidden email]> wrote:
> >
> >> Hi Sergey,
> >> Here is a nice example of the images I'm working with.  I have to
> measure
> >> the mean intensity of the whole blue area in the tip by drawing ROI
> around
> >> it (my supervisor doesn't like the tresholding)
> >>
> >> <
> >>
> http://imagej.1557.n6.nabble.com/file/n5002445/AUX1-GUS_1%C2%B5M_ACC_8.jpg
> >> >
> >>
> >> Best,
> >> Magi
> >>
> >>
> >>
> >>
> >>
> >> --
> >> View this message in context:
> >>
> http://imagej.1557.n6.nabble.com/GUS-staining-quantification-tp5002411p5002445.html
> >> Sent from the ImageJ mailing list archive at Nabble.com.
> >>
> >> --
> >> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
> >>
> >
> >
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

... [show rest of quote]

... [show rest of quote]

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

[Christian Goosmann-2]

Dear Magi,
in a brightfield image you are looking at densities not at intensities.
What you tried would be ok for fluorescence images where a present stain
(dye) produces a proportional intensity of blue light.
When you look at stains in a brightfield image you actually see the
highest intensity in all channels where you see white. Where your blue
stain is present, you see the same intensity in blue but less intensity
in the other channels mainly in red. So the information is indeed in the
red channel and it is inverted because the highest density of your dye
subtracts the most intensty out of your red channel. So if you want a
postive relation to your dye you need to invert the image and look at
the red channel. In a brightfield image you are looking at densities not
at intensities. I am sure there are functions in ImageJ that handle
densities rather than intensities. Look at functions like color
deconvolution etc. I am not too experienced with myself.
Just hope to point you in the right direction.
Christian

---
Christian Goosmann
Mikroskopie
Max-Planck-Institut für Infektionsbiologie
Campus Charité Mitte
Charitéplatz 1
10117 Berlin
Tel.: [phone]

magi78343 wrote:

> Dear all,
>
> I need to quantify the GUS-staining intensity (blue) of some digital images.
> As far as I know I first need to split the RGB channels. I am interested
> only in the blue intensity but the blue channel is the pilest one and the
> dark gray color corresponding to the dark blue on the color image remains in
> the red channel. Which channel should I use for measurement?
>
> Thank you in advance.
>
> Magi
>
>
>
> --
> View this message in context: http://imagej.1557.n6.nabble.com/GUS-staining-quantification-tp5002411.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]

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

[magi78343]

Hello again,
Thank you all for your advices.

As I try to avoid tresholding I decided to use the Red channel from the RGB stack. I don't want the results to be in gray values so I calibrated all images to a step-tablet (manual for the calibration: http://imagej.nih.gov/ij/docs/examples/calibration/). Now the results I get are in OD. Is the measure ment ok  like this or I should better use other way of calibration?

Best regards,
Magi

[G. Esteban Fernandez]

You can compute OD from the intensity value:  OD = log(1/transmittance),
where transmittance = intensity/max. possible intensity

-Esteban


On Wed, Mar 27, 2013 at 4:03 AM, magi78343 <[hidden email]> wrote:

> Hello again,
> Thank you all for your advices.
>
> As I try to avoid tresholding I decided to use the Red channel from the RGB
> stack. I don't want the results to be in gray values so I calibrated all
> images to a step-tablet (manual for the calibration:
> http://imagej.nih.gov/ij/docs/examples/calibration/). Now the results I
> get
> are in OD. Is the measure ment ok  like this or I should better use other
> way of calibration?
>
> Best regards,
> Magi
>
>
>
> --
> View this message in context:
> http://imagej.1557.n6.nabble.com/GUS-staining-quantification-tp5002411p5002455.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]

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