Using ImageJ for visceral fat in CT imaging



Dear Suni,

It should work on Mac OS as well. I have a colleague working on mac without problem.

I understand the error that you get but it shouldn’t happen, this is really puzzling.

Can we continue this disscussion by email ? (
I will need a sample of your images to try to reproduce your bug and eventually use some screen sharing service to show me how this bug occurs.

Since this software is really on the first version I expect to have some bugs to fix but I need to be able to duplicate the problem to fix it.

Best regards,



Dear Suni,

The problem might be related to a stupid mistake, I think I uploaded a wrong version on my Git,

Please try with this jar :

(Take care of replacing the old jar in Fiji and do not stay with 2 copy of the same jar)

In case of problem write me an email and we will investigate further.

Best regards,



Hi Salim

Still no luck with the new jar that you sent me. When I try to see the soft tissue or fat segmentation, the image is just black.

Will email you and send you the image that I am using

Thank you for all your help, it is truly appreciated



Hello, I have the same problem as you. I also need to get the fat areas from the CT image. I get the image (dicom format) from the CT scan. However, the threshold cannot get to the negative value. The type of the image is 8 bit and I don’t know how to adjust to 16 bit.
Any help will be thankful!


hello,i have the same question. The CT image I get is dicom format. However, the threshold cannot be negative value. And i find that the type of the discom format is 8bit. Is this the question?
Thank you very much!


Sorry but I can’t help you with DICOM-problems.

I guess that you need to know what’s the exact original format of your images and how these images are imported to/by ImageJ (plain ImageJ- or BioFormats-Import). In any case bipolar images need a special treatment when being imported. I don’t belive that bipolar CT-images should be converted to 8bit because they usually contain much more than 256 gray-values (usually ±1000 Hounsfield Units).




Thank you so much! I will check these images and I hope to find out the problem.