Abstract: The objective of this project is to create a publicly available electronic neuroanatomy resource that will advance neuroscience research and spread neuroanatomical knowledge. A large number of publicly available MRI brain scans will be labeled based on 60+ neuroanatomical regions of interest. Using semiautomatic software, technicians with extensive training will apply precisely defined measurement methods to make outlines around regions so that the entire brain is identified in every scan. We will then combine all of these results into a new kind of computerized atlas that can be used to 1) locate and identify brain regions in other scans, and 2) to interactively learn brain anatomy. We will make all of our results available to the public for scientific and educational purposes. The first part of this project has been to develop the comprehensive neuroanatomical labeling protocol that we will use to complete the project. We seek to compare and harmonize existing methods as much as possible and are looking for input and feedback. The next step will be to label 500-600 MRI brain scans. We have chosen the Open Access Series of Imaging Studies (OASIS) scans for the bulk of the work but will also analyze 152 normal adult brain MRI scans collected within the ICBM project (the ICBM152). After labeling the scans, we will assemble them into a Dynamic Probabilistic Atlas. This atlas will be generated on the fly from labeled scans based on criteria such as subject age, gender, handedness, etc. The database of labeled scans will also be used to create an MRI Brain Anatomy Instructor application. This interactive neuroanatomy teaching tool will allow “lesson” scripts to be created by authors and used by students. The main value of this project is to help understand and disseminate information about the variation of normal brain structure. It will also aide in the creation of automated MRI analysis methods. This project will provide an order of magnitude more manually labeled scans than are currently available.