Hi there,
I wonder if anyone knows how to use Fiji to copy a ROI from a MR T2-weightd image and paste it into a diffusion weighted image to include the same area of a lesion. The T2-weightd image and the diffusion weighted image have the same slice location and thickness but different FOV and matrix.
Please advise.
Thanks.
Lucas
I would recommend to add the ROI to the ROI Manager:
https://imagej.nih.gov/ij/docs/guide/146-30.html#fig:The-ROI-Manager
The stored ROI can be reenabled on any opened image.
In addition you can store all ROI Manager ROI’s to enable reproducible results.
Hi,
And I would add that you can also restore any ROI (it means that you can “duplicate” the active ROI of an image to another) with ctrl+Shift+e
Nico
Hi, Bio7,
Thank you for your kind suggestion. However, the locations of the ROIs on the two different images are not the same even though using ROI manger. It seems that I have to perform image registration, but I do not know which registration method I should use. Hope you can give me more suggestions.
Lucas
Hi, Bio7,
Thank you for your kind suggestion. However, the locations of the ROIs on the two different images are not the same even though using ROI manger. It seems that I have to perform image registration, but I do not know which registration method I should use. Hope you can give me more suggestions.
Lucas
Hi, Nico
Thank you for your kind suggestion. However, the locations of the ROIs on the two different images are not the same by using ctrl+Shift+e.
Lucas
I think It would be useful if you post example images here to see what you would like to archive and to help to understand your problem.
The following are two MR images of the same slice. You can find that the locations of the ROIs on them are different.
It took a while before I recognized that the stack dimensions are different (was this intentional?).
That’s why the ROI location is different in the second stack.
If you scale the second stack to 768x768 the locations should be equal
(However don’t do this because you change the original image data with scaling).
A simple macro correction (for the second stack) would be to correct the ROI location and size with the calculated different scale factor. Something like (select the first stack - adapt for yourself)
corr=640/768;//Calculate the correction! X,Y image size are equal!
roiManager("select", 0);//Select the first ROI!
Roi.getBounds(x,y,width,height);//Get the coordinates of the ROI!
run("Scale... ", "x="+corr+" y="+corr+" centered");//Scale the images!
setSelectionLocation(x*corr, y*corr);//Correct the location!
roiManager("add");//Add the correction to the Manager for the second stack!
Many thanks for your help. However, I have to deal with dozens of cases and I am not familiar with “maco”. Is there any more easier method to deal with my problem?
Simply paste and execute the given ImageJ macro code with Plugins->New->Macro on the first stack.
Execute with Macros->Run Macro
What do you mean by dozen of cases?
It would be helpful from the beginning if you describe you problem with more details (what do we see, which kind of data do you have, are all datasets different, what do you want to measure and why, etc.) so that the people in this forum can help you more efficiently.
Sorry, I did not describe my issue clearly. What I want to do is to measure the MR signal intensity of the same area of a tumor on various MR images. I will deal with MR images of more than forty cases. The datasets of these cases are different. I have to ensure that the ROIs on the images obtained from the same case include the same area of the tumor. For each case I have to do this correction, and therefore I prefer to use relative easier method. If simply adjusting the scale will not influence the measurement, I would like to use it.
Thank you for your kind help. Is there a built-in command in Fiji for getDimensions function?
If you mean in a script, it is the same. (Fiji is Just ImageJ 
)
If you mean in the menus, you can get the image size in Image>Properties…
Nico
Many thanks for your kind help. I will try it.
Lucas