Quantified methods to evaluate the CT - MRI alignment?

Hi, all
I’m doing brain T1-MRI and CT registration. Are there any Quantified Methods could help evaluate the quality of the registration, besides the visual judgement?


One that jumps to the top of my head is the Dice Score, which is a measure of volume alignment. It would not be able to judge surface alignment, if that is important to you.

As I know, If I want to use Dice, it means I already have a ground-truth-aligned CT.
While in my case, I used dipy to register CT and MRI, and I want to calculate some indexes like R2, or mutual information to show the difference between the original CT and aligned CT, to tell us the quality of the registeration.