engleski

Image-based measurements of intracerebral brain hemorrhage

A methodology for 3-D image-based measurements of human spontaneous intracerebral brain hemorrhage (ICH) is described in this work. The underlying medical problem is to perform quantitative measurements of the ICH region and edema region from computed tomography (CT) images. The goal of the project is to measure parameters for a large number of patients having ICH and to correlate measured parameters to patient morbidity and mortality. In the initial step, a CT film containing brain images is scanned into a computer. This preporcessing phase is based on fuzzy C-means clustering and provides a parameter-free procedure for normalization of image brightness. In the subsequent step a hierarchical segmentation method is used for locating the ICH and edema regions. Expert system rules form a knowledge base about the structure of a CT brain image and about the basic brain anatomy. The resulting ICH region is used as input to the second phase of analysis which is segmentation of edema. The edema segmentation has shown to be a difficult task even for neuroradiologists, so the system is designed so that user interaction is required to select one of the offered edema segmentation results. After the segmentation is finished the interpolation is performed to obtain a 3-D binary representation of ICH and edema regions and measure their volume and other parameters. The obtained 3-D volumes are used to perform the volume measurements and 3-D visualization of the ICH and edema regions.

computed tomography; image segmentation

nije evidentirano