The use of mutual information for rigid medical image registration: a review

Mutual information (MI) is a popular entropy-based similarity measure used in the medical imaging field for multimodal registration. Stemming from information theory, this measure generally outperforms most other intensity-based measures in multimodal applications as it only assumes a statistical dependence between images. This measure was introduced in the computer vision field by two separate and independent groups in 1995, using two slightly different formulations. This paper provides an introduction to this measure and its use in rigid medical image registration. A look at the extensions proposed to the original measure is also provided. These were developed to improve the robustness of the measure and to avoid certain cases when maximising MI does not lead to the correct spatial alignment.