threshold
Locked
 
|
Hi,
Can I manipulate with the image brightness before analysis using threshold? I am trying to determine brown area stained with DAB using image J, i tried to use threshold and color deconvolution. my images a dark can i use photoshop to adjust background or adjust brightness (using levels or curves) ? or that will change the results? Regards,Sarah M. Mosaad, M.Sc. -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
|
|
Hi Sarah,
Just noticed that no one seems to have replied to your question. This is probably because we aren't sure what "results" you are referring to and what you mean by "threshold". If you are measuring areas, then it's OK to use some processing to make the segmentation easier as long as you don't start to bring up some areas which were not really stained. If you were measuring intensity for fluorescence images, then you shouldn't use any processing which changes the pixel values. However, we know that measuring intensity/density of DAB staining is not acceptable. This has been discussed many times on the listserver. Please read the previous thread here: http://imagej.1557.x6.nabble.com/DAB-quantification-td5006159.html <http://imagej.1557.x6.nabble.com/DAB-quantification-td5006159.html> in particular the comments by Gabriel Landini in reference to DAB quantification. Kind regards, 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 Sarah Mohamed Sent: Tuesday, 16 September 2014 10:15 p.m. To: [hidden email]<mailto:[hidden email]> Subject: threshold Hi, Can I manipulate with the image brightness before analysis using threshold? I am trying to determine brown area stained with DAB using image J, i tried to use threshold and color deconvolution. my images a dark can i use photoshop to adjust background or adjust brightness (using levels or curves) ? or that will change the results? Regards,Sarah M. Mosaad, M.Sc. -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
|
|
Dear Jacqui,
thanks for your reply, i appreciate it. In your opinion, which stain can be used (to avoid the DAB issue with quantification) to quantify a marker in treated and untreated samples? Kind Regards, Sarah M. Mosaad, M.Sc. Department of Pharmacology & Toxicology, School of Pharmacy, Suez Canal University, Egypt. On Sunday, October 12, 2014 10:06 PM, Jacqui Ross <[hidden email]> wrote: Hi Sarah, Just noticed that no one seems to have replied to your question. This is probably because we aren't sure what "results" you are referring to and what you mean by "threshold". If you are measuring areas, then it's OK to use some processing to make the segmentation easier as long as you don't start to bring up some areas which were not really stained. If you were measuring intensity for fluorescence images, then you shouldn't use any processing which changes the pixel values. However, we know that measuring intensity/density of DAB staining is not acceptable. This has been discussed many times on the listserver. Please read the previous thread here: http://imagej.1557.x6.nabble.com/DAB-quantification-td5006159.html <http://imagej.1557.x6.nabble.com/DAB-quantification-td5006159.html> in particular the comments by Gabriel Landini in reference to DAB quantification. Kind regards, 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 Sarah Mohamed Sent: Tuesday, 16 September 2014 10:15 p.m. To: [hidden email]<mailto:[hidden email]> Subject: threshold Hi, Can I manipulate with the image brightness before analysis using threshold? I am trying to determine brown area stained with DAB using image J, i tried to use threshold and color deconvolution. my images a dark can i use photoshop to adjust background or adjust brightness (using levels or curves) ? or that will change the results? Regards,Sarah M. Mosaad, M.Sc. -- 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 |
|
|
Hi Sarah,
You can still quantify DAB staining but you need to measure parameters such as cell number, area, area fraction, distribution pattern, etc. Often area fraction can give you the data you are looking for. You just shouldn't measure intensities. One of the issues not mentioned in the previous discussion that I referred you to, is the need to set up the microscope perfectly every time for Koehler illumination, lamp temperature, etc. just to get consistent images. A lot of people don't do this correctly. If I was measuring absolute differences in protein expression, I'd probably be using biochemical techniques, e.g. Western blot. For relative differences in intensity, you could use immunohistochemistry with a fluorescent tag. However, there are many difficulties along that path as well, e.g. photobleaching, etc. and you need to select your antigen target carefully. Kind regards, 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 Sarah Mohamed Sent: Sunday, 19 October 2014 7:19 a.m. To: [hidden email] Subject: Re: threshold Dear Jacqui, thanks for your reply, i appreciate it. In your opinion, which stain can be used (to avoid the DAB issue with quantification) to quantify a marker in treated and untreated samples? Kind Regards, Sarah M. Mosaad, M.Sc. Department of Pharmacology & Toxicology, School of Pharmacy, Suez Canal University, Egypt. On Sunday, October 12, 2014 10:06 PM, Jacqui Ross <[hidden email]> wrote: Hi Sarah, Just noticed that no one seems to have replied to your question. This is probably because we aren't sure what "results" you are referring to and what you mean by "threshold". If you are measuring areas, then it's OK to use some processing to make the segmentation easier as long as you don't start to bring up some areas which were not really stained. If you were measuring intensity for fluorescence images, then you shouldn't use any processing which changes the pixel values. However, we know that measuring intensity/density of DAB staining is not acceptable. This has been discussed many times on the listserver. Please read the previous thread here: http://imagej.1557.x6.nabble.com/DAB-quantification-td5006159.html <http://imagej.1557.x6.nabble.com/DAB-quantification-td5006159.html> in particular the comments by Gabriel Landini in reference to DAB quantification. Kind regards, 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 Sarah Mohamed Sent: Tuesday, 16 September 2014 10:15 p.m. To: [hidden email]<mailto:[hidden email]> Subject: threshold Hi, Can I manipulate with the image brightness before analysis using threshold? I am trying to determine brown area stained with DAB using image J, i tried to use threshold and color deconvolution. my images a dark can i use photoshop to adjust background or adjust brightness (using levels or curves) ? or that will change the results? Regards,Sarah M. Mosaad, M.Sc. -- 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 -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
|
|
Dear Jacqui,
I would like to ask you about threshold method for determining area of brown stained cells. Should i mark the dark background and also the limit to threshold in setting measurment? And if not what would be the difference? i find that in some references they dont mark the dark background and others do.Also, Should the area value be the same if using threshold method or color deconvolution?Thanks in advance.Regards, Sarah On Monday, October 20, 2014 5:52 AM, Jacqui Ross <[hidden email]> wrote: Hi Sarah, You can still quantify DAB staining but you need to measure parameters such as cell number, area, area fraction, distribution pattern, etc. Often area fraction can give you the data you are looking for. You just shouldn't measure intensities. One of the issues not mentioned in the previous discussion that I referred you to, is the need to set up the microscope perfectly every time for Koehler illumination, lamp temperature, etc. just to get consistent images. A lot of people don't do this correctly. If I was measuring absolute differences in protein expression, I'd probably be using biochemical techniques, e.g. Western blot. For relative differences in intensity, you could use immunohistochemistry with a fluorescent tag. However, there are many difficulties along that path as well, e.g. photobleaching, etc. and you need to select your antigen target carefully. Kind regards, 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 Sarah Mohamed Sent: Sunday, 19 October 2014 7:19 a.m. To: [hidden email] Subject: Re: threshold Dear Jacqui, thanks for your reply, i appreciate it. In your opinion, which stain can be used (to avoid the DAB issue with quantification) to quantify a marker in treated and untreated samples? Kind Regards, Sarah M. Mosaad, M.Sc. Department of Pharmacology & Toxicology, School of Pharmacy, Suez Canal University, Egypt. On Sunday, October 12, 2014 10:06 PM, Jacqui Ross <[hidden email]> wrote: Hi Sarah, Just noticed that no one seems to have replied to your question. This is probably because we aren't sure what "results" you are referring to and what you mean by "threshold". If you are measuring areas, then it's OK to use some processing to make the segmentation easier as long as you don't start to bring up some areas which were not really stained. If you were measuring intensity for fluorescence images, then you shouldn't use any processing which changes the pixel values. However, we know that measuring intensity/density of DAB staining is not acceptable. This has been discussed many times on the listserver. Please read the previous thread here: http://imagej.1557.x6.nabble.com/DAB-quantification-td5006159.html <http://imagej.1557.x6.nabble.com/DAB-quantification-td5006159.html> in particular the comments by Gabriel Landini in reference to DAB quantification. Kind regards, 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 Sarah Mohamed Sent: Tuesday, 16 September 2014 10:15 p.m. To: [hidden email]<mailto:[hidden email]> Subject: threshold Hi, Can I manipulate with the image brightness before analysis using threshold? I am trying to determine brown area stained with DAB using image J, i tried to use threshold and color deconvolution. my images a dark can i use photoshop to adjust background or adjust brightness (using levels or curves) ? or that will change the results? Regards,Sarah M. Mosaad, M.Sc. -- 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 -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
|
|
Dear Sarah,
If you are thresholding DAB staining, then the background should be light. So, you should not select Dark background. Dark background is used for fluorescence images which have a black background. Yes, you do need to select Limit to threshold in the Set Measurements window. If you use Colour deconvolution, you still need to use thresholding to define the area you want to measure. The method for segmentation does affect the area measurements so you need to decide which is best for your images. However, if each method is done consistently, i.e. selects the area of interest reliably, the area measurements should not be very different. If you have DAB/Haematoxylin, I find that the Colour Decon works very well as long as you have the microscope set up properly. If you only have DAB, then I wouldn't bother with using the Colour Decon as you only have one colour. In this case, I would either split the channels and work on the blue channel (best contrast for DAB in my opinion) or just convert the colour image to grayscale and use that. I hope this is helpful to you. Kind regards, 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 Sarah Mohamed Sent: Tuesday, 18 November 2014 6:31 a.m. To: [hidden email] Subject: Re: threshold Dear Jacqui, I would like to ask you about threshold method for determining area of brown stained cells. Should i mark the dark background and also the limit to threshold in setting measurment? And if not what would be the difference? i find that in some references they dont mark the dark background and others do.Also, Should the area value be the same if using threshold method or color deconvolution?Thanks in advance.Regards, Sarah On Monday, October 20, 2014 5:52 AM, Jacqui Ross <[hidden email]> wrote: Hi Sarah, You can still quantify DAB staining but you need to measure parameters such as cell number, area, area fraction, distribution pattern, etc. Often area fraction can give you the data you are looking for. You just shouldn't measure intensities. One of the issues not mentioned in the previous discussion that I referred you to, is the need to set up the microscope perfectly every time for Koehler illumination, lamp temperature, etc. just to get consistent images. A lot of people don't do this correctly. If I was measuring absolute differences in protein expression, I'd probably be using biochemical techniques, e.g. Western blot. For relative differences in intensity, you could use immunohistochemistry with a fluorescent tag. However, there are many difficulties along that path as well, e.g. photobleaching, etc. and you need to select your antigen target carefully. Kind regards, 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 Sarah Mohamed Sent: Sunday, 19 October 2014 7:19 a.m. To: [hidden email] Subject: Re: threshold Dear Jacqui, thanks for your reply, i appreciate it. In your opinion, which stain can be used (to avoid the DAB issue with quantification) to quantify a marker in treated and untreated samples? Kind Regards, Sarah M. Mosaad, M.Sc. Department of Pharmacology & Toxicology, School of Pharmacy, Suez Canal University, Egypt. On Sunday, October 12, 2014 10:06 PM, Jacqui Ross <[hidden email]> wrote: Hi Sarah, Just noticed that no one seems to have replied to your question. This is probably because we aren't sure what "results" you are referring to and what you mean by "threshold". If you are measuring areas, then it's OK to use some processing to make the segmentation easier as long as you don't start to bring up some areas which were not really stained. If you were measuring intensity for fluorescence images, then you shouldn't use any processing which changes the pixel values. However, we know that measuring intensity/density of DAB staining is not acceptable. This has been discussed many times on the listserver. Please read the previous thread here: http://imagej.1557.x6.nabble.com/DAB-quantification-td5006159.html <http://imagej.1557.x6.nabble.com/DAB-quantification-td5006159.html> in particular the comments by Gabriel Landini in reference to DAB quantification. Kind regards, 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 Sarah Mohamed Sent: Tuesday, 16 September 2014 10:15 p.m. To: [hidden email]<mailto:[hidden email]> Subject: threshold Hi, Can I manipulate with the image brightness before analysis using threshold? I am trying to determine brown area stained with DAB using image J, i tried to use threshold and color deconvolution. my images a dark can i use photoshop to adjust background or adjust brightness (using levels or curves) ? or that will change the results? Regards,Sarah M. Mosaad, M.Sc. -- 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 -- 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 |
|
|
Dear Jacqui,Thanks for these valuable information, it really helped.I would like to ask you can i use images with different magnification power and compare there brown area in threshold or color Dec methods? but i will set a scale before that.
i am asking as on power 40X the images were dark while on 20X the images are more like what i see under the microscope and in both cases the white balance is set on (Auto), so i have now images with 2 magnification power and dont know if i have to repeat the shoting for those with 40X. I will appreciate your advice. Kind Regards,Sarah On Tuesday, November 25, 2014 4:50 AM, Jacqui Ross <[hidden email]> wrote: Dear Sarah, If you are thresholding DAB staining, then the background should be light. So, you should not select Dark background. Dark background is used for fluorescence images which have a black background. Yes, you do need to select Limit to threshold in the Set Measurements window. If you use Colour deconvolution, you still need to use thresholding to define the area you want to measure. The method for segmentation does affect the area measurements so you need to decide which is best for your images. However, if each method is done consistently, i.e. selects the area of interest reliably, the area measurements should not be very different. If you have DAB/Haematoxylin, I find that the Colour Decon works very well as long as you have the microscope set up properly. If you only have DAB, then I wouldn't bother with using the Colour Decon as you only have one colour. In this case, I would either split the channels and work on the blue channel (best contrast for DAB in my opinion) or just convert the colour image to grayscale and use that. I hope this is helpful to you. Kind regards, 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 Sarah Mohamed Sent: Tuesday, 18 November 2014 6:31 a.m. To: [hidden email] Subject: Re: threshold Dear Jacqui, I would like to ask you about threshold method for determining area of brown stained cells. Should i mark the dark background and also the limit to threshold in setting measurment? And if not what would be the difference? i find that in some references they dont mark the dark background and others do.Also, Should the area value be the same if using threshold method or color deconvolution?Thanks in advance.Regards, Sarah On Monday, October 20, 2014 5:52 AM, Jacqui Ross <[hidden email]> wrote: Hi Sarah, You can still quantify DAB staining but you need to measure parameters such as cell number, area, area fraction, distribution pattern, etc. Often area fraction can give you the data you are looking for. You just shouldn't measure intensities. One of the issues not mentioned in the previous discussion that I referred you to, is the need to set up the microscope perfectly every time for Koehler illumination, lamp temperature, etc. just to get consistent images. A lot of people don't do this correctly. If I was measuring absolute differences in protein expression, I'd probably be using biochemical techniques, e.g. Western blot. For relative differences in intensity, you could use immunohistochemistry with a fluorescent tag. However, there are many difficulties along that path as well, e.g. photobleaching, etc. and you need to select your antigen target carefully. Kind regards, 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 Sarah Mohamed Sent: Sunday, 19 October 2014 7:19 a.m. To: [hidden email] Subject: Re: threshold Dear Jacqui, thanks for your reply, i appreciate it. In your opinion, which stain can be used (to avoid the DAB issue with quantification) to quantify a marker in treated and untreated samples? Kind Regards, Sarah M. Mosaad, M.Sc. Department of Pharmacology & Toxicology, School of Pharmacy, Suez Canal University, Egypt. On Sunday, October 12, 2014 10:06 PM, Jacqui Ross <[hidden email]> wrote: Hi Sarah, Just noticed that no one seems to have replied to your question. This is probably because we aren't sure what "results" you are referring to and what you mean by "threshold". If you are measuring areas, then it's OK to use some processing to make the segmentation easier as long as you don't start to bring up some areas which were not really stained. If you were measuring intensity for fluorescence images, then you shouldn't use any processing which changes the pixel values. However, we know that measuring intensity/density of DAB staining is not acceptable. This has been discussed many times on the listserver. Please read the previous thread here: http://imagej.1557.x6.nabble.com/DAB-quantification-td5006159.html <http://imagej.1557.x6.nabble.com/DAB-quantification-td5006159.html> in particular the comments by Gabriel Landini in reference to DAB quantification. Kind regards, 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 Sarah Mohamed Sent: Tuesday, 16 September 2014 10:15 p.m. To: [hidden email]<mailto:[hidden email]> Subject: threshold Hi, Can I manipulate with the image brightness before analysis using threshold? I am trying to determine brown area stained with DAB using image J, i tried to use threshold and color deconvolution. my images a dark can i use photoshop to adjust background or adjust brightness (using levels or curves) ? or that will change the results? Regards,Sarah M. Mosaad, M.Sc. -- 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 -- 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 -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
|
|
Dear Sarah,
All of the images should be taken with the same resolution, both optical (objective lens) and digital (image size in pixels x pixels). Since you have DAB staining, you can go back and take the images again at the same magnification. Did you set up the microscope for Koehler illumination? If the images were dark at 40x, I would normally assume that the condenser position hasn't been set correctly (height and centering) and you probably have the condenser aperture set too small. Did you set the lamp temperature to 3200K? If you don't know what Koehler illumination is, let me know what microscope you are using and I'll send you some instructions off-list. It's important to do this microscope set-up correctly. Also, if you don't do this, your images won't be very good quality and may not be acceptable for publication. Kind regards, 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 Sarah Mohamed Sent: Thursday, 27 November 2014 12:45 p.m. To: [hidden email] Subject: Re: threshold Dear Jacqui,Thanks for these valuable information, it really helped.I would like to ask you can i use images with different magnification power and compare there brown area in threshold or color Dec methods? but i will set a scale before that. i am asking as on power 40X the images were dark while on 20X the images are more like what i see under the microscope and in both cases the white balance is set on (Auto), so i have now images with 2 magnification power and dont know if i have to repeat the shoting for those with 40X. I will appreciate your advice. Kind Regards,Sarah On Tuesday, November 25, 2014 4:50 AM, Jacqui Ross <[hidden email]> wrote: Dear Sarah, If you are thresholding DAB staining, then the background should be light. So, you should not select Dark background. Dark background is used for fluorescence images which have a black background. Yes, you do need to select Limit to threshold in the Set Measurements window. If you use Colour deconvolution, you still need to use thresholding to define the area you want to measure. The method for segmentation does affect the area measurements so you need to decide which is best for your images. However, if each method is done consistently, i.e. selects the area of interest reliably, the area measurements should not be very different. If you have DAB/Haematoxylin, I find that the Colour Decon works very well as long as you have the microscope set up properly. If you only have DAB, then I wouldn't bother with using the Colour Decon as you only have one colour. In this case, I would either split the channels and work on the blue channel (best contrast for DAB in my opinion) or just convert the colour image to grayscale and use that. I hope this is helpful to you. Kind regards, 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 Sarah Mohamed Sent: Tuesday, 18 November 2014 6:31 a.m. To: [hidden email] Subject: Re: threshold Dear Jacqui, I would like to ask you about threshold method for determining area of brown stained cells. Should i mark the dark background and also the limit to threshold in setting measurment? And if not what would be the difference? i find that in some references they dont mark the dark background and others do.Also, Should the area value be the same if using threshold method or color deconvolution?Thanks in advance.Regards, Sarah On Monday, October 20, 2014 5:52 AM, Jacqui Ross <[hidden email]> wrote: Hi Sarah, You can still quantify DAB staining but you need to measure parameters such as cell number, area, area fraction, distribution pattern, etc. Often area fraction can give you the data you are looking for. You just shouldn't measure intensities. One of the issues not mentioned in the previous discussion that I referred you to, is the need to set up the microscope perfectly every time for Koehler illumination, lamp temperature, etc. just to get consistent images. A lot of people don't do this correctly. If I was measuring absolute differences in protein expression, I'd probably be using biochemical techniques, e.g. Western blot. For relative differences in intensity, you could use immunohistochemistry with a fluorescent tag. However, there are many difficulties along that path as well, e.g. photobleaching, etc. and you need to select your antigen target carefully. Kind regards, 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 Sarah Mohamed Sent: Sunday, 19 October 2014 7:19 a.m. To: [hidden email] Subject: Re: threshold Dear Jacqui, thanks for your reply, i appreciate it. In your opinion, which stain can be used (to avoid the DAB issue with quantification) to quantify a marker in treated and untreated samples? Kind Regards, Sarah M. Mosaad, M.Sc. Department of Pharmacology & Toxicology, School of Pharmacy, Suez Canal University, Egypt. On Sunday, October 12, 2014 10:06 PM, Jacqui Ross <[hidden email]> wrote: Hi Sarah, Just noticed that no one seems to have replied to your question. This is probably because we aren't sure what "results" you are referring to and what you mean by "threshold". If you are measuring areas, then it's OK to use some processing to make the segmentation easier as long as you don't start to bring up some areas which were not really stained. If you were measuring intensity for fluorescence images, then you shouldn't use any processing which changes the pixel values. However, we know that measuring intensity/density of DAB staining is not acceptable. This has been discussed many times on the listserver. Please read the previous thread here: http://imagej.1557.x6.nabble.com/DAB-quantification-td5006159.html <http://imagej.1557.x6.nabble.com/DAB-quantification-td5006159.html> in particular the comments by Gabriel Landini in reference to DAB quantification. Kind regards, 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 Sarah Mohamed Sent: Tuesday, 16 September 2014 10:15 p.m. To: [hidden email]<mailto:[hidden email]> Subject: threshold Hi, Can I manipulate with the image brightness before analysis using threshold? I am trying to determine brown area stained with DAB using image J, i tried to use threshold and color deconvolution. my images a dark can i use photoshop to adjust background or adjust brightness (using levels or curves) ? or that will change the results? Regards,Sarah M. Mosaad, M.Sc. -- 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 -- 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 -- 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 |