Mutual information based image registration with application to 3D medical brain imagery

An image fusion algorithm accepts two or more images of the same region and produces an image with higher information content. The first step toward image fusion is a precise alignment of the images involved, such that the corresponding pixels/voxels in the two images/volumes represent the same physical point of the common region. This task is usually referred to as image registration in the literature. When the two images are acquired from different types of imaging sensors, this process is called multimodality image registration. Multimodality image registration has become an important research topic because of its great value in a variety of applications. For medical image analysis, an image showing functional and metabolic activity such as single photon emission computed tomography (SPECT), positron emission tomography (PET), and magnetic resonance spectroscopy (MRS), is often registered to an image which shows anatomical structures such as magnetic resonance image (MRI), computed tomography (CT), and ultrasound. These registered multimodality images are fused which, in turn, lead to improved diagnosis, better surgical planning, more accurate radiation therapy, and many other medical benefits.