Segmentation and features extraction techniques, with applications to biomedical images

In order to obtain a 3-D reconstruction of the hippocampus from a volumetric MRI head study, it is necessary to separate it not only from the surrounding white matter, but also from contiguous areas of gray matter. At present it is necessary for a physician to manually segment the hippocampus on each slice of the volume in order to obtain such a reconstruction. The authors propose a novel technique by which a computer may make an unsupervised identification of a given structure through a series of images, even if that structure includes so-called false contours or missing contours. Applications include 3-D reconstruction of difficult-to-segment regions of the brain and abdomen, and volumetric measurements of organs from series of 2-D images.