Cell counting
Locked
 
|
Hello everybody!
We want to count the brown and blue stained cell nuclei in order to calculate the proliferation rate (percentage of proliferative cells). Brown stands for proliferative, blue stands for non-proliferative. We have a lot of biopsies so we really would like to use ImageJ to do the cell counting. Two pictures are attached to show examples of two different tissues. The first one is at 20x magnitude, the second picture shows 40x magnitude. Would it be possible to use ImageJ here? best regards Jenny -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
33870-10 No.1.jpg (100K) |
|
Hi Jenny,
I'm relatively new to imagej and I see some problems with it, regarding the subjectivity of it. But I'd try to separate the color channels (although I don't know if thats possible with your sort of image (I use immunofluorescence of cell cultures and it's possible) Then I would download the color pixel counter. Afterwards you could calculate all coloured pixels / mean pixels of one cell to get the number of cells But as I said. I'm quite new to imagej |
|
|
Hi Jenny
Under the "Image->Adjust" menu there is a "color threshold..." plugin. This brings up a dialog that gives you the means to try thresholds in different color spaces. There is even a button to export to a macro. There is also a plugin called 'Color Inspector 3D'. This displays the pixel values in a 3D color cube. It also has the option to use different color spaces. Once you have something that seems to work on a few representative images you could record it as a macro. That's not always straight forward but the list can probably help you with questions about that. On Tue, Mar 31, 2015 at 1:48 PM, veraelisabeth <[hidden email]> wrote: > Hi Jenny, > > I'm relatively new to imagej and I see some problems with it, regarding the > subjectivity of it. But I'd try to separate the color channels (although I > don't know if thats possible with your sort of image (I use > immunofluorescence of cell cultures and it's possible) > > Then I would download the color pixel counter. Afterwards you could > calculate all coloured pixels / mean pixels of one cell to get the number > of > cells > > But as I said. I'm quite new to imagej > > > > -- > View this message in context: > http://imagej.1557.x6.nabble.com/Cell-counting-tp5012266p5012273.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 |
|
|
In reply to this post by veraelisabeth
Dear all,
I am not an expert with ImageJ, but I would like to contribute to the topic. Before counting the cells, what I would do is: 1. *Correct the background illumination* (see this LINK <http://correct background illumination>, there are ways so you need to see what is available and suits you). 2. Use *Colour Deconvolution* (*see this LINK <http://www.mecourse.com/landinig/software/cdeconv/cdeconv.html>*) or *Trainable Weka Segmentation* (*see this LINK <https://www.youtube.com/watch?v=8yfBHiGufFE>*) to isolate the cells that interest you. There are also other means, but see what it works for you. 3. Use ImageJ's built-in tool to count/analyse the cells. An other advice is to make sure the staining protocols, image acquisition and image analysis is consistent so there is no room for major errors. Hope it helps, Best regards, Andrei Stefan 2015-03-31 18:48 GMT+01:00 veraelisabeth <[hidden email]>: > Hi Jenny, > > I'm relatively new to imagej and I see some problems with it, regarding the > subjectivity of it. But I'd try to separate the color channels (although I > don't know if thats possible with your sort of image (I use > immunofluorescence of cell cultures and it's possible) > > Then I would download the color pixel counter. Afterwards you could > calculate all coloured pixels / mean pixels of one cell to get the number > of > cells > > But as I said. I'm quite new to imagej > > > > -- > View this message in context: > http://imagej.1557.x6.nabble.com/Cell-counting-tp5012266p5012273.html > Sent from the ImageJ mailing list archive at Nabble.com. > > -- > ImageJ mailing list: http://imagej.nih.gov/ij/list.html > -- Andrei Cătălin ȘTEFAN, MRCVS, DVM, MVSc E-mail: [hidden email] -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
|
|
Somehow I didn't added the link... But here it is:
1. *Correct the background illumination* (*see this LINK <http://imagejdocu.tudor.lu/doku.php?id=howto:working:how_to_correct_background_illumination_in_brightfield_microscopy>* ). 2015-03-31 19:29 GMT+01:00 Andrei Catalin Stefan < [hidden email]>: > Dear all, > > I am not an expert with ImageJ, but I would like to contribute to the > topic. > Before counting the cells, what I would do is: > > 1. *Correct the background illumination* (see this LINK, there are ways > so you need to see what is available and suits you). > 2. Use *Colour Deconvolution* (*see this LINK > <http://www.mecourse.com/landinig/software/cdeconv/cdeconv.html>*) or *Trainable > Weka Segmentation* (*see this LINK > <https://www.youtube.com/watch?v=8yfBHiGufFE>*) to isolate the cells that > interest you. There are also other means, but see what it works for you. > 3. Use ImageJ's built-in tool to count/analyse the cells. > > An other advice is to make sure the staining protocols, image acquisition > and image analysis is consistent so there is no room for major errors. > > Hope it helps, > > > > Best regards, > Andrei Stefan > > 2015-03-31 18:48 GMT+01:00 veraelisabeth <[hidden email]>: > >> Hi Jenny, >> >> I'm relatively new to imagej and I see some problems with it, regarding >> the >> subjectivity of it. But I'd try to separate the color channels (although I >> don't know if thats possible with your sort of image (I use >> immunofluorescence of cell cultures and it's possible) >> >> Then I would download the color pixel counter. Afterwards you could >> calculate all coloured pixels / mean pixels of one cell to get the number >> of >> cells >> >> But as I said. I'm quite new to imagej >> >> >> >> -- >> View this message in context: >> http://imagej.1557.x6.nabble.com/Cell-counting-tp5012266p5012273.html >> Sent from the ImageJ mailing list archive at Nabble.com. >> >> -- >> ImageJ mailing list: http://imagej.nih.gov/ij/list.html >> > > > > -- > > Andrei Cătălin ȘTEFAN, MRCVS, DVM, MVSc > > E-mail: [hidden email] > > -- Andrei Cătălin ȘTEFAN, MRCVS, DVM, MVSc E-mail: [hidden email] -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
|
|
Hi Jenny,
Just like to add my support to Andrei's recommendation that you use the Colour Deconvolution plugin for your initial segmentation. I couldn't tell from your images what resolution they are but hopefully you have the originals that you can use for the analysis. There are some instructions on our website that you can refer to if you need further assistance in using the plugin. Take a look here: https://www.fmhs.auckland.ac.nz/en/sms/about/our-departments/biomedical-imaging-research-unit/image-processing-and-analysis/analysis-resources.html Cheers, Jacqui Jacqueline Ross Biomedical Imaging Microscopist Biomedical Imaging Research Unit School of Medical Sciences Faculty of Medical & Health Sciences The University of Auckland Private Bag 92019 Auckland 1142, NEW ZEALAND Tel: 64 9 923 7438 Fax: 64 9 373 7484 http://www.fmhs.auckland.ac.nz/sms/biru/ -----Original Message----- From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of Andrei Catalin Stefan Sent: Wednesday, 1 April 2015 8:58 a.m. To: [hidden email] Subject: Re: Cell counting Somehow I didn't added the link... But here it is: 1. *Correct the background illumination* (*see this LINK <http://imagejdocu.tudor.lu/doku.php?id=howto:working:how_to_correct_background_illumination_in_brightfield_microscopy>* ). 2015-03-31 19:29 GMT+01:00 Andrei Catalin Stefan < [hidden email]>: > Dear all, > > I am not an expert with ImageJ, but I would like to contribute to the > topic. > Before counting the cells, what I would do is: > > 1. *Correct the background illumination* (see this LINK, there are > ways so you need to see what is available and suits you). > 2. Use *Colour Deconvolution* (*see this LINK > <http://www.mecourse.com/landinig/software/cdeconv/cdeconv.html>*) or > *Trainable Weka Segmentation* (*see this LINK > <https://www.youtube.com/watch?v=8yfBHiGufFE>*) to isolate the cells > that interest you. There are also other means, but see what it works for you. > 3. Use ImageJ's built-in tool to count/analyse the cells. > > An other advice is to make sure the staining protocols, image > acquisition and image analysis is consistent so there is no room for major errors. > > Hope it helps, > > > > Best regards, > Andrei Stefan > > 2015-03-31 18:48 GMT+01:00 veraelisabeth <[hidden email]>: > >> Hi Jenny, >> >> I'm relatively new to imagej and I see some problems with it, >> regarding the subjectivity of it. But I'd try to separate the color >> channels (although I don't know if thats possible with your sort of >> image (I use immunofluorescence of cell cultures and it's possible) >> >> Then I would download the color pixel counter. Afterwards you could >> calculate all coloured pixels / mean pixels of one cell to get the >> number of cells >> >> But as I said. I'm quite new to imagej >> >> >> >> -- >> View this message in context: >> http://imagej.1557.x6.nabble.com/Cell-counting-tp5012266p5012273.html >> Sent from the ImageJ mailing list archive at Nabble.com. >> >> -- >> ImageJ mailing list: http://imagej.nih.gov/ij/list.html >> > > > > -- > > Andrei Cătălin ȘTEFAN, MRCVS, DVM, MVSc > > E-mail: [hidden email] > > -- Andrei Cătălin ȘTEFAN, MRCVS, DVM, MVSc E-mail: [hidden email] -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
|
|
Thank you so much! I will try your suggestions and I hope we can manage
somehow. If there are some further questions from my side I will contact you again. Thanks for once. Good day to you all Jenny 2015-04-01 1:14 GMT+02:00 Jacqui Ross <[hidden email]>: > Hi Jenny, > > Just like to add my support to Andrei's recommendation that you use the > Colour Deconvolution plugin for your initial segmentation. I couldn't tell > from your images what resolution they are but hopefully you have the > originals that you can use for the analysis. > > There are some instructions on our website that you can refer to if you > need further assistance in using the plugin. > > Take a look here: > https://www.fmhs.auckland.ac.nz/en/sms/about/our-departments/biomedical-imaging-research-unit/image-processing-and-analysis/analysis-resources.html > > Cheers, > > Jacqui > > Jacqueline Ross > Biomedical Imaging Microscopist > Biomedical Imaging Research Unit > School of Medical Sciences > Faculty of Medical & Health Sciences > The University of Auckland > Private Bag 92019 > Auckland 1142, NEW ZEALAND > > Tel: 64 9 923 7438 > Fax: 64 9 373 7484 > > http://www.fmhs.auckland.ac.nz/sms/biru/ > > > -----Original Message----- > From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of > Andrei Catalin Stefan > Sent: Wednesday, 1 April 2015 8:58 a.m. > To: [hidden email] > Subject: Re: Cell counting > > Somehow I didn't added the link... But here it is: > > 1. *Correct the background illumination* (*see this LINK > < > http://imagejdocu.tudor.lu/doku.php?id=howto:working:how_to_correct_background_illumination_in_brightfield_microscopy > >* > ). > > 2015-03-31 19:29 GMT+01:00 Andrei Catalin Stefan < > [hidden email]>: > > > Dear all, > > > > I am not an expert with ImageJ, but I would like to contribute to the > > topic. > > Before counting the cells, what I would do is: > > > > 1. *Correct the background illumination* (see this LINK, there are > > ways so you need to see what is available and suits you). > > 2. Use *Colour Deconvolution* (*see this LINK > > <http://www.mecourse.com/landinig/software/cdeconv/cdeconv.html>*) or > > *Trainable Weka Segmentation* (*see this LINK > > <https://www.youtube.com/watch?v=8yfBHiGufFE>*) to isolate the cells > > that interest you. There are also other means, but see what it works for > you. > > 3. Use ImageJ's built-in tool to count/analyse the cells. > > > > An other advice is to make sure the staining protocols, image > > acquisition and image analysis is consistent so there is no room for > major errors. > > > > Hope it helps, > > > > > > > > Best regards, > > Andrei Stefan > > > > 2015-03-31 18:48 GMT+01:00 veraelisabeth <[hidden email]>: > > > >> Hi Jenny, > >> > >> I'm relatively new to imagej and I see some problems with it, > >> regarding the subjectivity of it. But I'd try to separate the color > >> channels (although I don't know if thats possible with your sort of > >> image (I use immunofluorescence of cell cultures and it's possible) > >> > >> Then I would download the color pixel counter. Afterwards you could > >> calculate all coloured pixels / mean pixels of one cell to get the > >> number of cells > >> > >> But as I said. I'm quite new to imagej > >> > >> > >> > >> -- > >> View this message in context: > >> http://imagej.1557.x6.nabble.com/Cell-counting-tp5012266p5012273.html > >> Sent from the ImageJ mailing list archive at Nabble.com. > >> > >> -- > >> ImageJ mailing list: http://imagej.nih.gov/ij/list.html > >> > > > > > > > > -- > > > > Andrei Cătălin ȘTEFAN, MRCVS, DVM, MVSc > > > > E-mail: [hidden email] > > > > > > > -- > > Andrei Cătălin ȘTEFAN, MRCVS, DVM, MVSc > > E-mail: [hidden email] > > -- > ImageJ mailing list: http://imagej.nih.gov/ij/list.html > > -- > ImageJ mailing list: http://imagej.nih.gov/ij/list.html > -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
«
Return to ImageJ
|
1 view|%1 views
| Free forum by Nabble | Edit this page |