ANTsX
Negative mutual information
Hi,
After applying a SyN registration, I wanted to know the similarity of the two images. Therefore I used ants.image_similarity(fixed, registered, metric_type='MattesMutualInformation', sampling_strategy='None').
This resulted in a negative value for Mutual information, although this is not possible from theory and the images are actually quite well registered. Can someone comment on this?
Another option for me would be to calculate the mutual information myself, which is not a problem. But I was just curious why this output was negative.
Best, Ilva
All metrics minimize - so we return negative MI
On Mon, Jan 10, 2022 at 3:51 AM IlvaHou55 @.***> wrote:
Hi,
After applying a SyN registration, I wanted to know the similarity of the two images. Therefore I used ants.image_similarity(fixed, registered, metric_type='MattesMutualInformation', sampling_strategy='None'). This resulted in a negative value for Mutual information, although this is not possible from theory and the images are actually quite well registered. Can someone comment on this? Another option for me would be to calculate the mutual information myself, which is not a problem. But I was just curious why this output was negative.
Best, Ilva
— Reply to this email directly, view it on GitHub https://github.com/ANTsX/ANTsPy/issues/315, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACPE7S32DCJPAOW7SGQNV3UVKMZTANCNFSM5LTCZKZQ . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
You are receiving this because you are subscribed to this thread.Message ID: @.***>
--
brian
