We propose a center dedicated to research and development of computerized image analysis technology for us with samples produced by modern neuroanatomical techniques. Included among these techniques are receptor and radiolabeled tracer autoradiography which reveal detailed functional and distributional information about neural activity and mechanisms. Center research falls into two broad categories: parcellation, or rapid and reproducible identification of the boundaries of subareas in histological images, and quantification, or the measurement of the relative or absolute values of underlying substance values from optical images of these regions. The parcellation project utilizes expert system technology for knowledge based image analysis. Quantification projects include the development of a computerized system for the determination of local thickness of variation, of computer techniques for the detection of local variations within brain regions, and for the quantification of autoradiograms. The center will operate service facilities that will supply image analysis software for microcomputer image analysis systems used in neuroscience, and will also provide access to image analysis hardware for researchers in the Northeast Corridor and throughout the country. The image analysis hardware and software assists in the acquisition of accurate and precise information from CNS samples, focusing on the detection, quantification, and mapping of the experimental effects seen in autoradiographic and other images and in the accurate collation of different images (e.g., autoradiographic and histological images). The center will also disseminate information about computerized image analysis methods for neuroscience by training researchers, publishing papers, and offering seminars and short courses.
Showing the most recent 10 out of 22 publications
