Ok: I have pictures of heart tissue with a staining of WGA, SMA and Dapi. I am looking for arteries. Now I am trying to measure the lumen of this arteries. But to avoid drawing a area by hand for each artery I tried to mark the black areas of the picture by using Image J-Image-adjust-colorThreshold.
I highlight stained reagions in the picture by saturating them. Now i select everything that is not red(saturated). Obviously there are much more areas marked, than just the lumen of the artery.
Is there a way to determine the "circle" around the lumen as a ROI?
Thank you in advance
Hey Simon, I am not sure if I understand it right. From your description, I guess "crop" works. If you don't need to look at other regions, you could crop only the artery and do the same color threshold adjustment as before.
If you have multiple arteries to measure on the same image, another way is to use the ROI manager, which is under "Analyze - Tools - ROI Manager ". You could choose multiple ROIs here and measure them together.
Also if you are still using ImageJ, consider to update to Fiji which is the newer version of ImageJ.
Hope it helps. Good luck.
Good morning Penny. Thank you for your answer, but I think that doesn't work for me.
I have to measure the area of the lumen and if I set a ROI around the artery or the lumen of it, Image J shows me the area of the rectangle I drawed and not of the lumen, even if I mark it like in Picture 2. If I just measure the area of the marked reagions (Picture 2) it gives me the whole area of dark reagions in the Picture, wether or not there is a ROI around the lumen of the artery.
To crop the artery out would be possible, but if I do it for every Image with normaly 2 or 3 arteries in it, the time I need, would be equally Long (Iguess), as if I encircle every lumen by hand (in Zenblue).
Won't work without cropping. You have too many regions with no staining. I would just crop, make the image 8 or 16 bit gray, threshold, create selection based using the binary thresholded image, make inverse, and measure the area.
Hi Simon,
Here is what MIPAR could do to automatically find just the Lumen feature. The gist of its applied logic is:
I used your posted "red threshold" image to build the recipe since that is all I had, but the recipe could be easily modified to work from the raw image. I've attached the recipe file -- just open your first attached image in MIPAR's Image Processor and load the recipe!
It is often when folks think something cannot be done with ImageJ, or find it too difficult, that they are delighted to find MIPAR!
Also -- always here for any questions.
Cheers,
John
- Badges
- Science method