Informatics Algorithms for Neural Circuitry Ultrastructure

The overall objective of this proposal is the creation of informatics algorithms for analysis of neural circuitry ultrastructure. Determining the detailed connections in brain circuits is a fundamental unsolved problem in neuroscience. Understanding this circuitry will enable brain scientists to confirm or refute existing models, develop new ones, and come closer to an understanding of how the brain works. Clinical imaging of prematurely born infants has identified disruptions of the normal development of gray matter and white matter as the major risk factors associated with the frequent occurrence of long-term adverse neurodevelopmental outcomes. The development of technology for imaging and analyzing the maturation and connectivity of neurons, axons and synapses will provide critical new insight into the mechanisms of injury that cause poor outcomes, and facilitate the search for successful interventions. Recent advances in electron microscopy imaging now provide the capability to image the ultrastructure of the brain at unprecedented resolution with extremely large volumes of interest. Recent work is continuing to increase the resolution of voxels to smaller sizes, acquiring even more voxels, while enabling automated capture of larger regions. However, these advances in image acquisition have not yet been matched by advances in algorithms and implementations that will be capable of enabling the analysis of neural circuitry. Recent reviews have identified the development of appropriate informatics tools as the primary requirement for new success in understanding the connectivity of the brain. The specific aims of this proposal are to facilitate the analysis and interpretation of neural ultra-structure by creating 3D volumes of neural ultrastructure from 2D images and large 2D images of neural ultrastructure from 2D camera tiles, and to perform segmentation and detection of neural ultrastructure. The research to achieve each of these specific aims involves the development, implementation and evaluation of novel informatics algorithms especially designed to meet the requirements of high resolution large data acquisition electron microscopy of neural ultrastructure.