This application requests image-processing instrumentation to upgrade and update a Shared Neuroscience Image-Processing Facility, serving the needs of a group of strongly funded neuroscientist and associated research personnel who utilize quantitative autoradiographic and non- autoradiographic image-processing in their research. These efforts are becoming critically compromised as a result of several factors: 1) Obsolescence and Malfunction: Our image-processors have repeatedly broken down: 1 or 2 units is current beyond repair; and the manufacturer has been out of business for 8 years. Our densitometer is over 10 years old and extraordinarily expensive to maintain. 2) Computational and Workstation Limitations: The central processor speed of our present system is 30-50-fold slower that current-generation workstations- a prohibitive constraint as regards both program development and image- analysis. For example, three-dimensional autoradiographic processing of a single brain currently requires 3 days of CPU time. Our two current image-processing workstations, even if both were functional, would be too few to accommodate the increasing demands of our multiple users. The proposed image-processing system will serve seven senior-level independently funded investigators along with computer analysts, junior investigators, and research associates and assistants. This group is actively supported by 3 major NIH Program Projects/Centers in the fields of cerebral vascular disease, neurotrauma (head injury) and spinal cord injury; and by 5 individual R01/R03 NIII research grants (including a Jacob Javits Investigatorship) in the areas of blood-brain barrier function, and receptor binding and kinetics. A vigorous software- development effort is underway to implement novel algorithms for two- and three-dimensional image-alignment and reconstruction; double-label autoradiographic strategies for sequential assessment of two differing functional/metabolic states; and co-registration of replicate studies so as to derive mean and variance images and to do arithmetic manipulations on multiple image files.
