Collaborative, Distributed Database for Brain Viewing

Distributed Collaborative Database for Brain Viewing This project will develop a distributed database system and the collaborative viewing tools necessary to overcome the difficulties inherent in imageintensive neurobiological research that is accomplished by collaborations between geographically separated researchers. Images containing morphological information either are, or have recently become, a primary form of data in many fields of biological research. For an individual laboratory, the iceberg metaphor is true. Published images are the tip, whereas the ever- expanding collection of raw image files stored on hard disks and removable media is the hidden body of the iceberg. Neuroscience is particularly image intensive. From the perspective of a single laboratory, managing the magnitude of the data collected in recent years is daunting. Neuroscientists often wish to compare images of brains from experimental groups with those from control animals. It is oftentimes a challenge to identify and understand the spatial relations of a brain region identified by a staining or labeling procedure. Furthermore, collaborations among researchers in the same or different institutions are impeded by lack of image viewing tools. This goals of this project present challenges for computer scientists, including the design and development of: (i) additional file header information that represents biological variables and is generated at the microscope during the capturing of the image using speech-to-text technology; (ii) viewing tools needed for collections of images that give perspective, while retaining the pristine data captured from the digital camera; (iii) multipleresolution image-handling techniques integrated with a DBMS to meet data intensive image requirements; (iv) collaboration support for on-going research discussion centered around images, across time and space boundaries; (v) techniques for integrating audio and text within the same discussion. This research wi ll yield two kinds of results: a) artifacts, software and databases, that will be demonstrated and made available to others, and b) reports in the literature on studies of a variety of research questions on the impact of these on neuroscientists' activity.