One method of aligning two medical images is to extract features from the images, then compute the best alignment of the features [7]. This approach depends greatly on the ability to automatically and accurately extract reliable image features. In general, methods of feature extraction such as intensity thresholding or edge detection do not work well on medical scans, due to non linear intensity biases and highly textured structures. Without the ability to accurately localize corresponding features in the images, alignment in this manner is difficult.
A second registration method uses fiducial markers attached to a patient throughout the various acquisitions. If the markers can easily be located in the images, the volumes can be registered by computing the best alignment of the corresponding fiducials [8,12]. The main drawback of this method is that the markers must remain attached to the patient throughout all image acquisitions.