Roger Woods - US grants

The objectives of this proposal are: i) to optimize and validate positron emission tomography (PET) activation techniques as a method of functional brain mapping in single subjects, and 2) to apply these techniques to better understand processes of sensory-motor integration in the human. PET studies will use radioactively labeled water as a means of measuring regional changes in blood flow in response to behavioral tasks. The methodologic aspects of this proposal will focus on improvements in signal-to-noise ratios by using better imaging protocols, development of more accurate statistical models for analyzing data from individual subjects, and independent validation of PET findings by comparing the results with those obtained using other methods of functional evaluation. The objective of these methodologic investigations is to develop functional brain imaging as a valid, rigorous tool for the study of normal and abnormal brain physiology in individual subjects. The objective of the neuroscientific component of the proposal is to use this tool to better understand sensory-motor integration in the human. Initial studies will map sensory-motor processes in normal subjects as a function of the sensory input modality and the motor output modality. Participation of the parietal lobe in sensory-motor tasks will be of particular interest, but the demonstration of one or more widespread cortical networks involved in sensory-motor integration is anticipated. Further studies will examine the -lateralization of sensory-motor processing in normal subjects to look for evidence of hemispheric asymmetries. During the final two years of the proposal, patients with sensory-motor deficits and/or with lesions in brain structures normally involved in sensory-motor integration will be studied. Multimodal integration of clinical observations, functional imaging studies, anatomic imaging studies, and electrophysiologic measurements will be utilized in normal subjects and in patients to further clarify the relationship between brain structure and function.

The award will support the development, testing and evaluation of algorithms to enable the functional and structural mapping of the brain between different modalities, subjects and developmental stages. The algorithms will spatially deform data from varied sources so that they become comparable. The algorithms will be applied to positioning, warping, interpolating and parameterizing of 3D brain data. The PI will model brain using volume and surface based approaches, then transform it using methods ranging from rigid repositioning to nonlinear, local deformations. Transformation algorithms will first be measured for accuracy, reliability and validity via experiments using simulations and previously acquired data. The goal is to develop algorithms that are biologically valid as well as numerically accurate. This work will complement the Human Brain Project research and will enable the integration of independent yet complementary brain information from different sources and projects.

Specific Aim 1. Develop methods for extracting and visualizing differences in brain shape using information obtained from volume-based nonlinear warping algorithms such as AIR. Specific Aim 2. Develop and validate statistical methods to allow brain shape descriptors derived from nonlinear warping algorithms to be compared across groups of subjects. Specific Aim 3. Develop and validate methods for using shape descriptors derived from nonlinear warping algorithms to characterize developmental morphometric changes over time. Specific Aim 4. Develop and validate a statistical framework for comparing the temporal dynamics of morphometric changes across different groups of subjects. Specific Aim 5. Cross-validate and integrate the methods developed here with methods based on Surface Parameterization (Project 1) and other comparable or compatible approaches.

DESCRIPTION (provided by applicant): Innovations in imaging technologies and in the field of computational neuroscience have resulted in the generation of massive amounts of image and statistical data concerning the human brain. Additionally, recent advancements in the acquisition of genetic and other biomedical data have created novel interdisciplinary studies in the field of neuroscience and biomedical informatics. These advances have resulted in the need for high performance computing and novel investigative paradigms for the meaningful analysis of the available data. The Laboratory of Neuro Imaging has been a frontrunner in the adoption of cutting-edge technology to understand dynamic changes such as the development and degeneration of the human brain in health and disease. Our group has gained worldwide recognition for the development of innovative research methodologies, computational algorithms and high-order mathematical approaches to the investigation of brain registration, the analysis of variance between and within populations, and the visualization of these data. These techniques, however, now must not only accommodate four dimensions, as ongoing projects at LONI and elsewhere generate time-varying, multidimensional statistical fields but also be able to reasonably process data that are increasingly more complex. We now routinely apply these computationally demanding methods to population studies in order to achieve sensitivity to potentially subtle differences. In response to these computational challenges, a group of neuro-, biomedical and computer scientists with common interests and computational needs have come together to seek funding to greatly increase the storage capacity and bandwidth of a dedicated shared high performance computing facility. The increasing storage demands placed on this system by four-dimensional and high angular resolution imaging sets, used for intricate surface extraction, nonlinear warping and the multi-modal integration of complex is surfaces with data from disparate sources have clearly identified storage bandwidth limitations and gross storage capacity concerns that limit the progress of large imaging studies. The requested storage upgrade would eliminate capacity bandwidth contention and allow for the optimal utilization of the computational resource by LONI investigators and collaborators. An administrative plan is already in place by which the equipment can be managed equitably. Technical and management personnel also are part of the funded group of participants. Ongoing collaborations and the common programmatic requirements will enable sharing of computer code, analytic procedures, computational strategies and infrastructural capabilities. The requested instrument will enhance the productivity of ongoing computational biomedical research at LONI and collaborating sites in schizophrenia, HIV/AIDS and Alzheimer's disease, among others, and foster the development of leading edge technology and applications for a diverse array of collaborators and multidisciplinary investigators.

DESCRIPTION (provided by applicant): Relative to many mammalian species, the brains of humans and other primates are immature at birth and undergo extensive postnatal changes. These include gray matter changes in neurotransmitter receptor levels and extensive white matter changes in myelination, which continue even into adulthood. Different brain regions mature at different rates, so a complete characterization of postnatal development requires an anatomic context, ideally at high spatial resolution. Regional differences in maturational processes are thought to play a major role in the timing of normal developmental milestones and are also hypothesized to define critical temporal periods of vulnerability during which brain injury may lead to specific patterns of deficits. Critical periods in postnatal brain development has been identified in an NIMH National Advisory Mental Health Council Workgroup report as a high priority area for accelerating translational research in mental illness. Human developmental studies of neurotransmitter receptors are limited in terms of time points and regions sampled and in terms of the number of receptor types characterized, a situation unlikely to change given the difficulties of obtaining suitable post-mortem postnatal human tissues and the radiation exposures required for in vivo measurements. Post-mortem human studies of myelination are available, but provide only broad anatomic generalizations. Likewise, in vivo human myelination studies using magnetic resonance (MR) imaging have limited resolution, are typically based on fast acquisitions that potentially confound myelin with other signals and are complicated by the fact that the positions of individual tracts within white matter change as a result of myelination. Though not subject to the same fundamental limitations, available non-human primate studies of postnatal brain development are nonetheless undersampled both temporally and spatially, and comprehensive studies of multiple neurotransmitter receptors are not available. We propose here in a model system to characterize, at microscopic resolution and at six different postnatal developmental time points, myelination of white matter and the concentrations of 19 different major neurotransmitter receptor subtypes in gray matter. These will be placed in the context of micro-anatomic features (cytoarchitectonics in gray matter and tracts defined at ultrahigh (60-100 5m) resolution in white matter using 3D polarized light imaging). These microscopic studies will be supplemented by antecedent in vivo imaging of the same specimens, including structural, high angular resolution diffusion (HARDI), and resting state functional magnetic resonance (MR) imaging at 3 Tesla. Post-mortem MR scanning of 90 banked hemispheres at fifteen developmental time points at 7 Tesla will provide finer temporal sampling and allow more detailed characterization of individual variability of myelination (evaluated using quantitative MR T2 relaxometry) and white matter tract localization (evaluated using HARDI). All results will be integrated into a user-friendly, atlas- based on-line resource with links to other on-line human and non-human primate developmental resources.

? DESCRIPTION (provided by applicant): The major technological and analytical advances in human brain imaging achieved as part of the Human Connectome Projects (HCP) enable examination of structural and functional brain connectivity at unprecedented levels of spatial and temporal resolution. This information is proving invaluable for enhancing our understanding of normative variation in young adult brain connectivity. It is now timely to use the tools and analytical approaches developed by the HCP to understand how structural and functional wiring of the brain changes during the aging process. Using state-of-the art HCP imaging approaches will allow investigators to push our currently limited understanding of normative brain aging to new levels. We propose an effort involving a consortium of five sites (Massachusetts General Hospital, University of California at Los Angeles, University of Minnesota, Washington University in St. Louis, and Oxford University), with extensive complementary expertise in human brain imaging and aging and including many investigators associated with the original adult and pilot lifespan HCP efforts. This synergistic integration of advances from the MGH and WU-MINN-OXFORD HCPs with cutting-edge expertise in aging provides an unprecedented opportunity to advance our understanding of the normative changes in human brain connectivity with aging. Aim 1 will be to optimize existing HCP Lifespan Pilot project protocols to respect practical constraints in studying adults over a wide age range, including the very old (80+ years). Aim 2 will be to collect high quality neuroimaging, behavioral, and other datasets on 1200 individuals in the age range of 36 - 100+ years, using matched protocols across sites. This will enable robust cross-sectional analyses of age-related changes in network properties including metrics of connectivity, network integrity, response properties during tasks, and behavior. Aim 3 will be to collect and analyze longitudinal data on a subset of 300 individuals in three understudied and scientifically interesting groups: ages 36-44 (when late maturational and early aging processes may co-occur); ages 45-59 (perimenopausal, when rapid hormonal changes can affect cognition and the brain); and ages 80 - 100+ (the `very old', whose brains may reflect a `healthy survivor' state). The information gained relating to these important periods will enhance our understanding of how important phenomena such as hormonal changes affect the brain and will provide insights into factors that enable cognitively intact function into advanced aging. Aim 4 will capitalize on our success in sharing data in the Human Connectome Project (HCP), and will use these established tools, platforms, and procedures to make this data publicly available through the Connectome Coordination Facility.