Re: Query about the use of the MRI Wound Assay Tool

Posted by Volker Baecker on Feb 23, 2017; 10:06am
URL: http://imagej.273.s1.nabble.com/Re-Query-about-the-use-of-the-MRI-Wound-Assay-Tool-tp5018176.html

Hello Charmaine,
you can see the result I get on the image in the attachment. I'd say it
doesn't look to bad. But you have to see if it works for more than one
image. What I did:

Install the macro toolset from the attachment by copying it into the
folder macros>toolsets of your *FIJI* installation.

Install the plugins FeatureJ and MorphoLibJ.

Restart FIJI and select the toolset.

1. Right-click on the m-button of the toolset and set the parameter
values as following:

radius close: 8
radius open: 5
radius big close: 22
min. size: 100000
sigma for coherency: 1.4
window size for coherency: 3

2. Process>Subtract Background  with radius 15 and light background
selected.

3. Edit>Invert

4. Image>Adjust Brightness & Contrast, Auto

5. Press the m-button of the toolset

6. File>Revert to see the result on your original image.

Best regards,
Volker Bäcker

Charmaine Lo:

> Dear Dr. Baecker,
>
>
> My name is Charmaine Lo, an honours research student at the
> University of Newcastle, Australia.  I've recently come across your
> MRI Wound Healing Tool and have tried to implement it to our lab
> pictures to get more accurate and faster measurements of our wounds.
>
>
> The MRI tool works correctly using the sample pictures given on your
> website.  Unfortunately, the tool doesn't work for our own lab
> pictures.
>
>
> I have attached three pictures: one that is original (8bit), and one
> with the MRI measurement, and one with the MRI measurement after
> converting the original to 16bit.
>
>
> I already implemented your advice on your site and tried adjusting
> the picture to no effect. I was wondering whether you may have any
> advice you can give?
>
>
> I fully appreciate your time in helping us, and for making this tool
> and helping so many researchers worldwide.
>
>
> Kind regards,
>
> Charmaine Lo
>
>
>

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html