I want to calculate collagen volume fraction from Masson's-Trichrome stained slide images. The ImagePro software cannot be installed due to some technical difficulties. Can ImageJ software be used instead?
ImageJ can measure volume, surface area, area, perimeter, diameter, center of mass, and on and on. Post a sample image and we can guide you much better. Also I recommend the Fiji distribution of ImageJ as it comes loaded with plugins and features a great update system. Also, ImageJ and Fiji are free open source
Hello Kristopher
As per your suggestion I am posting a sample image here. This is a Masson'-Trichrome stained Cardiac tissue section image. The regions stained blue indicate collagen deposition. Can you please guide me how to calculate collagen volume (%) in this entire section? Looking forward for your help......
Nachiket explained it well. My only addition is that the image without any scale properties will have measures in units of pixels. Area in pixels^2, Perimeter/Length in pixels. If you capture the image from a scope that retains the scale info in the raw image, then the measures should be in those units (most likely um^2 or um).
If the image is 2D, then the only volume you can measure is an extrapolation. Multiply the thickness of the specimen by the area results to get volume. Better is to use the integrated density mentioned by Nachiket. This measure include pixel intensity, or the amount of collagen stain present. More stain -> more collagen. Coupled with area, the intensity measure incorporate the amount of collagen within the measured area. This then becomes a better measure than the volume for the purposes of comparing quantity of collagen among tissues. If however, you wish to compare the area occupied by some collagen amount above some defined criteria (threshold), then area is a good measure to use.
Experiment with the threshold controls. Try out the different automatic threshold methods to see which works best for your situation. Also try the color model selection and turning on/off the pass option for each of the three information channels. In the end you are trying to establish a good objective method which doesn't require a highly trained expert to segment, nor does it include the subjectivity of the person performing the segmentation. This method can then be incorporated into simple macro routines for high-throughput analysis. The much higher numbers of images analyzed can outweigh minor accuracy issues (n=1000 @ 95% is better than n=6 @ 100%). Also, any inaccuracry will be objective and systematic, not subjective to an experts opinion.
Thanks a lot Nachiket and Kristopher. Will definitely try out these methods and hope can solve it out........................
Best of luck. If you run into problems, post them here and we will try to help you along
Yes, area is in square units. The units are set in the Properties... command. So if you have set the scale then the area will be in these units. A cool tip is that after setting the unit and width, height, depth of the pixel you can go back and change the unit and it will automatically change the size values. E.g., set units: um, height: 1.5, width: 1.5, depth: 1.5 then go back to units and change it to mm. The 1.5 values will now show 0.0015.
You can use color deconvolution plugin to measure which color or region you interested in.
with ImageJ.
With all your suggestions, I have finally worked out the collagen volume fractions among my various experimental groups and the data are in accordance to my senior's data who actually did it with ImagePro software.
Thanks a lot guys.........................................
Hi aramita can u send me how u r analysing images using image software we are facing problems in getting reproducibility due to adjustment in threshold level
I know this is an old thread, but if you still need help, I would defintely recommend you visit www.MIPAR.us. I think it would work perfectly for your needs.
It is very powerful and intuitive 2D/3D image analysis software written by scientists/end-users and is about to be released as a free trial on December 14th. You can sign up on the site as well as submit images/datasets to test.
There are many image analysis options available, but I think you may just find MIPAR to be something special if you give it a try!
Anyone has the YouTube step by step to show how to quantify blue area in trichrome image?
Thanks
Youping
If you can post the image I can send you a MIPAR recipe file (used in MIPAR at http://MIPAR.us) to do it automatically. It is also educational since you can see or edit the entire non-destructive sequence.
Amarita, I am sorry if the software did not work out for you. Happy to help with a custom recipe or training anyway I can.
Cheers,
John
Thanks John.
here is one of my trichrome stain of heart section.
could you directly email to me, since I am no always on this web
my email address is:
Thank millions
youping
Hi Youping,
Thanks for posting. Here is the recipe file, recipe screenshot, and segmented results (overlaid on a grayscale version for easier visualization) I came up with. I will also send this to via email and we can discuss what exact measurements you needed performed on this image.
Cheers,
John
Yes, you can; I used it to quantify the degree of collagen deposition in lung tissue.
I did it as percentage of collagen/total tissue area in rat lung tissue slides.
Hi Mohamed abdelgied
Can you please tell us how to quantify the percentage of collagen/total tissue (In the case of lung tissue which has air) , How to exclude the air?
Thank you.
- Badges
- Science topic
- Similar topics
- Biological Science
- Histology