Multimodal 3-D image registration of MRI-SPECT volume images

We have developed a method to implement multimodal 3-D image registration using external markers, visible in both imaging modalities. This method was validated in a controlled mode using simulated images, so that various characteristics of the modalities (e.g. spatial resolution, slice thickness, pixel size) could be convolved with these images. In the present work we apply this method to real images of a given phantom acquired by MRI and SPECT.