Michael I. Miller - US grants

Physiological studies have now reached the point where the parameters of the broadband acoustic stimuli and the statistics of the random process auditory-nerve discharges which represent them may be understood. This suggests that questions concerning the representation of the complex acoustic cues fundamental to discrimination tasks can be resolved. For example, which of the broad array of auditory neurons carries statistically reliable phase-locked frequency information to the cochlear nucleus? Alternatively, how should cells in the central nervous system combine the phase-locked and place coded information across neurons for the frequency discrimination task or for the interaural phase discrimination required for sound localization. The proposed research addresses these questions by developing the Siebert/Gaumond stochastic model an validating it by accounting for the statistics of population responses to simple tone and speech stimuli. From the stochastic model, optimum coding strategies may be developed for the representation of the spectral and phase information fundamental to the discrimination of complex stimuli. The understanding of the coding of these stimuli which this work will provide is essential for the prediction of performance limits for normal as well as hearing impaired, and for the design of cochlear prostheses. This work should also provide a conceptual framework for guiding studies into the complex processing in the central nervous system. For example, optimum processor theory predicts the precise ways in which the information derived from the array of auditory-nerve fibers must converge onto task specific processing cells in higher nuclei so as to maintain the statistical integrity of the frequency and phase information required for the processing task. This work will also provide a new tool for the study of auditory nerve phenomena having to do with onset dynamic range effects, adaptation and the metabolic/physiologic properties of various classes of neurons. The joint maximum-likelihood estimation method for the generation of stimulus and recovery functions will provide a window into the previously unexaminable via spike discharge recordings transduction process between the stimulus related intracellular hair-cell potential and the action potential generation process.

This research involves studies in speech coding, image processing, and inference on stochastic processes. Recursive algorithms will be used for generation of maximum-likelihood and maximum-entropy estimates in problems for which the data are noisy and incomplete. Initially, the research applications will be predominantly in the areas of speech coding in the auditory system and image processing and spectral estimation for array processing.

The research deals with work in spectrum estimation for problems involving finite samples from stationary Gaussian stochastic processes. Emphasis is on the general method for achieving estimators with small bias and low variance. Theoretical work is in three areas: 1) Implementation via lattice filters, 2) Maximum-entropy extension, 3) Complex exponentials in noise. Investigations also involve multi-signal array processing, beamforming, and delay-doppler radar imaging along with mapping of these signal processing algorithms to parallel architectures. This project is one supported under the Creativity Awards for Undergraduate Engineering Students.

A hypercube parallel processor will be provided for researchers at Washington University for the Department of Electrical Engineering and Computer Science. This equipment is provided under the Instrumentation Grants for Research in Computer and Information Science and Engineering program. The research for which the equipment is to be used will be in the area(s) of: 1. Developing parallel hierarchical (circuit, gate, system) simulation system for use in computer aided design. 2. Research and implementation of a parallel iterative algorithm which performs maximum likelihood estimation of a multi-signal source in a noisy environment. 3. Developing software methodologies for concurrent system design and debugging.

The work proposed herein is towards the development of a comprehensive model for the analysis of signal transmission between the mammalian sensory receptor hair cell-afferent synapse and associated afferent fiber. A description is formulated in terms of interacting stochastic processes which are direct analogues of the physiological mechanisms of the mechanical/electrical operation of the synapse-afferent complex: vesicle recycle to synaptic body, vesicle transfer from synaptic body to docking sites, vesicle release via exocytosis, excitatory post-synaptic potential generation, and action potential generation. The model of the receptor synapse is based on the current experimental understanding of its molecular mechanisms of exocytosis allowing for a comprehensive study of the pre- and post-synaptic regulation of transmission in a systematic way. Such a model provides the vehicle for synthesis and prediction of the complex nonstationary responses of the synapse-afferent complex. A testimony to the model is in its predictive power. On the post-synaptic side, it predicts a direct link between the postsynaptic recovered discharge properties with the interaction between synaptic conductance and post-synaptic threshold voltage dynamics. It also predicts that the long- time constants associated with post-synaptic recovered discharge are linked to a yet to be identified long-time constant hyperpolarization in the post-synaptic afferent membrane, and it predicts the existence of the nonmonotonic discharge peak seen by various investigators in recovered response histograms at absolute dead-time. Simultaneously the model predicts a regime over which the Siebert/Gaumond discharge intensity product model holds. On the pre-synaptic side, the model predicts how the roles played by the synaptic stochastic processes, vesicle transfer from synaptic body to docking sites and vesicle exocytosis rate, determine the complex recovery and adaption rates observed during stimulation. The framework in which the receptor cell synapse is defined is general enough that it can be extended to more complex synaptic systems, such as that seen in the acousticolateral system in fish. This will allow for the direct modeling of afferents with multiple synapses and complex dendritic trees, revealing the mechanisms by which synaptic morphology directly determines afferent response dynamics. A model of synaptic transduction also provides researchers studying the regularity properties of neurons in the cochlear nucleus with realistic inputs with which to understand processing at higher levels. The major contribution is to provide an organized setting in which hypotheses concerning the receptor-synapse- afferent system can be tested. Viewing the synapse-afferent system as the concatenation of multiple interacting random processes, it allows for the direct inference from action potential data of the mechanisms of receptor cell transduction, inferences which are free from the distortive effects of pre- and post-synaptic adaptation. This can become a major tool for formulating questions about the roles played by the various components of the synapse and associated afferent fibers in encoding sensory information.

Abstract

PI: Donald J Geman

Title: ITR SMALL: Invariant Detection and Interpretation of Specific Objects in Image Data

The area of investigation is automated scene analysis. The main objective is to detect the appearance in image data of objects from a small repertoire. Two key liabilities in current methods are insufficient invariance, both photometric and geometric, and inefficient computation. To confront these difficulties, a unified statistical and computational framework is proposed which is based on a coarse-to-fine sequence of approximations to a full Bayesian model. Research topics include both algorithmic and mathematical issues arising in coarse-to-fine search, model selection and deformable shape analysis.

The interpretation of natural scenes is effortless for human beings but is the main challenge of artificial vision. This "semantic gap" has largely resisted any satisfying solution and impedes scientific and technological advances in many areas, including automated medical diagnosis, industrial automation, and effective security and surveillance. The general objective of this project is to design computer algorithms for detecting and interpreting certain objects appearing in still pictures in order to relieve humans of wearisome visual search tasks in medical imaging, law enforcement, industrial inspection and everyday life

DESCRIPTION (provided by applicant): One central issue in current functional MRI research is the problem of precisely localizing regions of activation and associating these regions with anatomical labels. Functional MRI data tend to have both a low signal-to-noise ratio and a low spatial resolution compared with conventional structural MRI data. There is also considerable biologically based individual variability in the shape of the brain that is a significant confounding variable in associating the activity seen in functional MRI with a specific brain region, particularly in the cortex. One partial solution to this problem of individual variability transforms individual functional MRI data to an atlas coordinate system in the hope that this transformation will increase the precision of structural-functional co-localization. The purpose of the proposed research is to characterize and quantify the reliability and validity of two distinct approaches for relating functional activation to the anatomical domain. The first approach, adopted by many researchers and exemplified by the SPM software, maps functional MRI data from multiple anatomical coordinate systems directly into an atlas coordinate system. The second approach maps the individual's functional data to that individual's own structural coordinates, to which high-dimensional transformations are applied carrying the structural and function information into the common atlas coordinate system. For decomposing the inherent variability and validity of this approach our own BrainWorks software will be examined and compared to commercially available BrainVoyager software for mapping the functional data to that individual's own structural domain and from the structural domain to an atlas coordinate system.

DESCRIPTION (provided by applicant): One central issue in current functional MRI research is the problem of precisely localizing regions of activation and associating these regions with anatomical labels. Functional MRI data tend to have both a low signal-to-noise ratio and a low spatial resolution compared with conventional structural MRI data. There is also considerable biologically based individual variability in the shape of the brain that is a significant confounding variable in associating the activity seen in functional MRI with a specific brain region, particularly in the cortex. One partial solution to this problem of individual variability transforms individual functional MRI data to an atlas coordinate system in the hope that this transformation will increase the precision of structural-functional co-localization. The purpose of the proposed research is to characterize and quantify the reliability and validity of two distinct approaches for relating functional activation to the anatomical domain. The first approach, adopted by many researchers and exemplified by the SPM software, maps functional MRI data from multiple anatomical coordinate systems directly into an atlas coordinate system. The second approach maps the individual's functional data to that individual's own structural coordinates, to which high-dimensional transformations are applied carrying the structural and function information into the common atlas coordinate system. For decomposing the inherent variability and validity of this approach our own BrainWorks software will be examined and compared to commercially available BrainVoyager software for mapping the functional data to that individual's own structural domain and from the structural domain to an atlas coordinate system.

This focused study proposes to analyze the structure induced on spaces of shapes by the action of groups of diffeomorphisms equipped with a right invariant metric. The project contains four components related to these spaces: their geometric analysis, the development of appropriate statistical methods, the required numerical analysis, and the application of the results to medical imaging and computational anatomy. The general framework is addressed by a new approach which in some sense formalizes the mechanics of shapes. Lie groups with right invariant metrics indeed are structures on which classical laws of mechanics can be shown to hold, and in particular the conservation of momentum along paths of minimal energy. It turns out that this momentum is a key to the representation and characterization of deformations in this context. It is albeit difficult to handle, because it is usually singular, as a measure, or a distribution on a singular support. This and the numerical difficulty it creates is probably one of the main challenges that we address in our study. Other important aspects are the study of the geometry such an approach induces on shape spaces, including a study of their curvatures, and the existence and stability of normal coordinates. This will be related to open issues in shape statistics, and applied in particular to biomedical imaging problems.

This approach is therefore designed to provide new tools for describing and analyzing shapes. Although shapes are prevalent in the outside world and in science, this is a difficult problem. For the human mind, there is an intuitive notion of what shapes are, why they differ or look alike, or when they present abnormalities with respect to ordinary observations. Sculpture is the art of rendering existing shapes, or creating new ones, and the fact that artists are still able to provide unambiguous instances of subjects through distorted or schematic representations is a strong indication of the robustness of the human shape recognition engine. However, an analytical description of a shape is much less obvious, and humans are much less efficient for this task, as if the understanding and recognition of forms work without an accurate extraction of their constituting components. We can recognize a squash from an eggplant or a pepper via a simple outline, and even provide a series of discriminative features which distinguish them, but it is much harder to phrase a verbal description of any of them, accurate enough, say for a painter to reproduce it. It is therefore not surprising that, for mathematics, shape description remains mostly a challenge. There are however very important applications which depend on progresses made in this field, one of them being the computerized analysis of biomedical shapes (computational anatomy), which analyzes the impact of diseases on shapes of organs, obtained from modern techniques of non-invasive 3D imagery. The last fifty years of research in computer vision has shown a amazingly large variety of points of view and techniques designed for this purpose: 2D or 3D sets they delineate (via either volume or boundary), moment-based features, medial axes or surfaces, null sets of polynomials, configurations of points of interest (landmarks), to name but a few. Yet, it does not seem that any of these methods has emerged as ideal, neither conceptually nor computationally, for describing shapes. An important aspect of our study will be to describe shapes with an indirect approach, from the way they can be deformed. This is not a new idea, and can be traced back to the seminal works of D'Arcy Thompson at the beginning of the 20th century, but its mathematical formalization and the design of practical algorithms is a comprehensive task, still offering many open problems, that the present group will try to address and convey to the scientific community.

Affective Disorders; Algorithms; Alzheimer; Alzheimer disease; Alzheimer sclerosis; Alzheimer syndrome; Alzheimer's; Alzheimer's Disease; Alzheimers Dementia; Alzheimers disease; Amentia; Ammon Horn; Anatomic; Anatomical Sciences; Anatomy; Apoplexy; Arts; Attention; Basal Ganglia; Basal Nuclei; Brain; Brain Diseases; Brain Disorders; Brain Mapping; Brain imaging; Brain region; CRISP; Cell Nucleus; Cerebral Ischemia; Cerebral Stroke; Cerebrovascular Apoplexy; Cerebrovascular Stroke; Cerebrovascular accident; Child Development Disorders; Childhood; Childhood Brain Neoplasm; Childhood Brain Tumor; Clinical; Cognitive; Cognitive Disturbance; Cognitive Impairment; Cognitive decline; Cognitive deficits; Cognitive function abnormal; Communities; Computer Retrieval of Information on Scientific Projects Database; Cornu Ammonis; Daily; Data; Deep; Dementia; Dementia, Alzheimer Type; Dementia, Primary Senile Degenerative; Dementia, Senile; Depression; Depth; Development; Developmental Disabilities; Diffusion; Diffusion MRI; Diffusion Magnetic Resonance Imaging; Diffusion Weighted MRI; Discipline; Disease; Disorder; Disturbance in cognition; Drug Formulations; Encephalon; Encephalon Diseases; Encephalons; Epilepsy; Epileptic Seizures; Epileptics; Fiber; Formulation; Formulations, Drug; Funding; Generalized Growth; Generations; Grant; Gray; Gray unit of radiation dose; Growth; Gyrus Hippocampi; Gyrus Parahippocampalis; Hippocampal Gyrus; Hippocampus; Hippocampus (Brain); Human; Human, General; Imagery; Imaging technology; Impaired cognition; Institution; Intracranial CNS Disorders; Intracranial Central Nervous System Disorders; Investigation; Investigators; Ischemia; Knowledge; Life; MR Imaging; MR Tomography; MRI; MS (Multiple Sclerosis); Magnetic Resonance; Magnetic Resonance Imaging; Magnetic Resonance Imaging Scan; Man (Taxonomy); Man, Modern; Maps; Measurement; Medical Imaging, Magnetic Resonance / Nuclear Magnetic Resonance; Memory; Mental Depression; Metabolic Diseases; Metabolic Disorder; Method LOINC Axis 6; Methodology; Methods; Metric; Mood Disorders; Multiple Sclerosis; NIH; NMR Imaging; NMR Tomography; National Institutes of Health; National Institutes of Health (U.S.); Neocortex; Nerve Cells; Nerve Degeneration; Nerve Unit; Nervous System, Brain; Neural Cell; Neuranatomies; Neuranatomy; Neuroanatomies; Neuroanatomy; Neurobiology; Neurocyte; Neuron Degeneration; Neurons; Normal Tissue; Normal tissue morphology; Nuclear Magnetic Resonance Imaging; Nucleus; Parahippocampal Gyrus; Pattern; Pediatric Neoplasm of the Brain; Pediatric Tumor of the Brain; Pediatrics; Physiologic; Physiological; Prefrontal Cortex; Primary Senile Degenerative Dementia; Process; Property; Property, LOINC Axis 2; Psychiatry; Range; Reading; Research; Research Personnel; Research Resources; Researchers; Resolution; Resources; Schizophrenia; Schizophrenic Disorders; Science of Anatomy; Sclerosis, Disseminated; Seizure Disorder; Shapes; Site; Source; Staging; Standards; Standards of Weights and Measures; Stroke; Structure; Surface; System; System, LOINC Axis 4; Thalamic structure; Thalamus; Thesaurismosis; Tissue Growth; Trauma; Trauma, Brain; Traumatic Brain Injury; Traumatic encephalopathy; United States National Institutes of Health; Vascular Accident, Brain; Visualization; Work; Zeugmatography; anatomy; biomed informatics; biomedical informatics; brain attack; brain visualization; cerebral vascular accident; cognitive dysfunction; cognitive loss; cognitively impaired; computerized data processing; data processing; data structure; dementia of the Alzheimer type; dementia praecox; design; designing; diffusion tensor imaging; disease/disorder; epilepsia; epileptiform; epileptogenic; frontal cortex; frontal lobe; gray matter; hippocampal; homotypical cortex; insular sclerosis; isocortex; metabolism disorder; neopallium; neural degeneration; neurobiological; neurodegeneration; neuronal; neuronal degeneration; ontogeny; pediatric; primary degenerative dementia; reconstruction; schizophrenic; senile dementia of the Alzheimer type; shape analysis; shape description; signal processing; size; software systems; stroke; substantia alba; substantia grisea; thalamic; tool; tool development; traumatic brain damage; tumor; white matter

DESCRIPTION (provided by applicant): The long term goal of Computational Anatomy (CA) is to create algorithmic tools that aid basic and clinical neuroscientists in the analysis of variability in anatomical structures at different scales. The difficulty is the complexity of anatomical substructures and the large variation across subjects. It is proposed to develop an open-source pipeline for 3D statistical shape analysis of anatomical variations from a population of anatomical structures. The overall aim is to integrate 3D Slicer application and ITK software library with the statistical shape analysis pipeline being disseminated by the Biomedical Informatics Research Network and thus enable the wider neuroimaging community to efficiently analyze anatomical variations in disease. The first aim is to standardize shape deformation vectors generated by several CA methods such as the Large Deformation Diffeomorphic Metric Mapping (LDDMM) developed at the Center for Imaging Science at Johns Hopkins University and the Finite Element Method for Deformable Registration (FEMDR) used in ITK. This will allow shape vectors to be used by both global metric classifier analysis in classifying diseased shapes and Gaussian Random Field (GRF) model analysis in localizing shape changes in disease. The two methods will be unified to provide a new metric classifier based on the data generated by GRF. In the final stage, hypothesis testing will be used to correlate global metric classification with localized shape changes. The second aim is to construct anatomical atlases needed for analysis of shape vectors. These atlases will be generated from segmented hippocampal and amygdala structures in already acquired populations of children, adolescents and young adults in neuroimaging studies of major depression disorder (MDD) at Washington University at St Louis. As a major public health burden, MDD provides the biological testbed for the pipeline from which probabilistic atlases will be generated. The third aim is to integrate the software libraries with the pipeline by leveraging the power and flexibility of the 3D Slicer software and ITK libraries developed by NA-MIC, Kitware and others. The fourth aim is to implement modules for visualization of the analysis of shape vectors in 3D Slicer. The fifth aim is to implement a stand-alone version of Medical Reality Markup Language (MRML) independent of 3D Slicer. This will allow for the propagation of MRML as a standard format for future neuroimaging applications. The shape analysis pipeline will be disseminated for use in neuroimaging studies of psychiatric disorders under the auspices of NA-MIC. PUBLIC HEALTH REVELANCE: This multidisciplinary, multi-institutional investigation, based on powerful computational anatomy and computer science software, has a strong potential to add significantly to the etiology of neurodevelopmental and neurodegeneration disorders. The driving biological motivation comes from complementary neuroimaging studies of early onset major depression disorder given the considerable public health burden of depression worldwide. The increased importance of early onset illness combined with the application to a population-based sample of twin pairs appears as an attractive model for statistical shape analysis software for the neuroscience community.

DESCRIPTION (provided by applicant): Huntington disease (HD) is a progressive, fatal, neurodegenerative disease, with movement disorder, psychiatric features, and cognitive decline. The neurodegeneration is regionally heterogeneous with preferential loss of striatal medium spiny neurons, but with significant atrophy in other regions. This leads to the question whether this pattern of regional degeneration is circuit related, reflecting the anatomic connections of the affected neurons, or by contrast is multifocal. To address this question, we will perform statistical shape analysis of basal ganglia and examine white matter structures connecting atrophied regions with affected cortical regions. We hypothesize that there will be heterogeneous atrophy in selected subcortical regions, and that shape analysis may detect some localized changes early, before overall volumes change significantly. We hypothesize that regional globus pallidus atrophy will correlate with specific local basal ganglia connections, but that regional striatal atrophy will not entirely correlate with connections predicted by regional cortical atrophy. We will also perform complementary analysis of white matter structures. Specifically Aim 1 will perform cross- sectional statistical shape analysis (caudate, putamen, thalamus, hippocampus, nucleus accumbens and globus pallidus) in 351 subjects with and without prodromal HD; Aim 2 will perform longitudinal shape analysis on specific subcortical gray matter regions (as listed above) for 351 subjects with scans at 2 time points; and Aim 3 will perform analysis of white matter structures in to determine whether the regions of striatum most affected receive projections from the regions of cortex most affected, and whether the regions of globus pallidus most affected receive projections from the regions of striatum most affected. PUBLIC HEALTH RELEVANCE: Specific sub-regions of atrophy identified in prodromal and early symptomatic Huntington's Disease (HD) will help determine whether neurodegeneration in HD follows a circuit-based pathway connecting brain structures (similar to prion disease, and as hypothesized for Alzheimer's and Parkinson's disease), or is multifocal. These data will be important for planning interventions, which directly target the brain and thus will be directly relevant for HD therapeutics.

DESCRIPTION (provided by applicant): The purpose of this grant is to support continued development and maintenance of MriStudio software developed in Johns Hopkins University. MriStudio is comprehensive software for MR image processing and analysis with emphasis on white matter anatomy. MriStudio consists of three modules, DtiStudio, DiffeoMap, and RoiEditor. DtiStudio was introduced in 2001 and remains one of the most widely used programs to process diffusion tensor imaging (DTI) data. DiffeoMap and RoiEditor were introduced in 2007, which provides a very unique environment to perform a cutting-edge image transformation and atlas-based automated image segmentation. What is especially unique about DiffeoMap is, because our advanced brain mapping algorithms are highly CPU and memory intensive, it adopts Cloud-type architecture, through which users can have access to our supercomputation resource. Currently, there are more than 6,500 registered uses. In this application, we propose to extend this service to the community through the following aims; Aim 1: Continued user support through training and dissemination Currently, two major channels of training and dissemination are through web-based resources (manuals and videos) and hand-on monthly 2-day tutorials. As the functionalities of MriStudio expand, we will continuously update the online materials and tutorials. Aim 2: Extension of the functionality Aim 2-1: Advanced diffusion MRI analysis package: Through the collaboration with Dr. Tournier, spherical harmonic decomvolution algorithm will be implemented. Aim 2-2: Automated and probabilistic tractography: We will incorporate a probabilistic tractography method based on dynamic programing and automate the ROI definition process. Aim 2-3: Quality control reporting: We will deploy a comprehensive and quantitative quality control reporting system, which is extremely important for automated analysis of large-scale studies. Aim 3: Cross-platform extension by adopting web-based interface. We will develop web-based Cloud computation service, which will eliminate the platform-dependence. Aim 4: Completion of XNAT-based solution, we will develop a server-based automated analysis pipeline that is linked to a research image database system, called XNAT. Aim 5: Deployment of multi-atlas-based brain segmentation algorithm, we will deploy our multi-atlas technology in our server and make them available for testing to users through the Cloud computation system.

DESCRIPTION (provided by applicant): IBM estimates that 30% of the entire data in the world is medical information. Medical images occupy a significant portion of medical records with approximately 100 million scans in US and growing every year. In addition, the data size from each scan steadfastly increases as the image resolution improves. These BigData are not structured and due to lack of standardized imaging protocols, they are highly heterogeneous with different spatial resolutions, contrasts, slice orientations, etc. In this project, we will deelop a technology to structure and search medical imaging information, which will make the past data available for education and evidence-based clinical decision-making. In this grant, we will focus on brain MRI, which comprises the largest portion of MRI data. The target community will be physicians who make decisions and the patients will be the ultimate beneficiaries. Currently, radiological image data are stored in clinical database called PACS. The image data in PACS are not structured. Consequently, once the diagnosis of a patient is completed, most of the data in PACS are currently discarded in the archive. Radiologists rely on their experience and education to reach medical decisions. This is a typical problem in medical practice that calls for objective evidence-based medicine. There are many ongoing attempts to structure the text fields of PACS, which include natural language processing of free-text radiological reports, clinical information, and diagnosis. In our approach, we propose to structure the image data, not text fields, to support direct search of images. Namely, physicians will submit an image of a new patient and search past images with similar anatomical phenotypes. Then, the clinical reports of the retrieved data will be compiled for a statistical report of the diagnosis and prognosis. We believe this image structuration is the key to unlock the vast amounts of information currently stored in PACS and use them for education and modern evidence-based medical decisions. The specific aims are; Objective 1: To develop and test the accuracy of high-throughput image structuration technologies Objective 2: To develop and test the image search engine Objective 3: Capacity Building Requirement: To develop prototype cloud system for data structuration / search services for research and educational purposes

IMAGING CORE - CORE C: ABSTRACT The Imaging Core (Core C) is responsible for all aspects of image acquisition and analysis. The aims of the Imaging Core are: (1) To facilitate the acquisition of new magnetic resonance imaging (MRI) scans and new Positron Emission Tomography scans, using Pittsburgh Compound B (PiB/PET), for the BIOCARD study participants, and to analyze the newly acquired scans with state-of-the-art methodology; (2) To continue the analysis of the previously acquired MRI scans with both semi-automated and automated methods; (3) To continue to work collaboratively with the other Cores to examine combinations of biomarkers that are predictive of progression and to examine the order and pattern of biomarker change during preclinical AD, including evaluating potential new biomarkers; (4) To share the image files and related clinical information with investigators in the field, via the BIOCARD website, using the approved BIOCARD Study data sharing procedures.

Project Summary Over the past decade, we have been building, parsing and wrangling systems for extracting neurodegeneration and neurodevelopment biomarkers from high-dimensional magnetic resonance (MR) imagery at 1 mm3 scale which are discriminating. At the same time, large and complex data sets and networks of segmented structures are becoming increasingly available to the research community such as Predict-HD, Track-HD, ADNI, and SchizConnect. Neuroscientists and clinicians are interested in tracking biomarkers which characterize rates of atrophy in anatomical networks, onset of or changepoint times of spread through the networks, and prediction of risk to conversion as determined by clinical symptoms. These wrangling and modeling methods are novel. Our biomarkers are extracted via brain mapping technologies based on diffeomorphometry, the study of morphological change via diffeomorphic tracking of anatomical coordinate systems at the sub millimeter scale. Like stereology, diffeomorphometry discovers high-dimensional features signalling neurodegeneration and neurodevelopment via tight integration of random field based statistical methods via large deviation empirical probability estimators calculated via high-dimensional permutation testing. Family-wise rates are calculated for group comparisons, and have been advanced changepoint modelling allowing us to explicitly estimate the spread of progression of anatomical feature change through the networked structures associated to neurodegeneration - Alzheimer's Disease (AD) and Huntingdon's Disease (HD) and neurodevelopment - Schizophrenia (SZ) and Attention Deficit and Hyperactive Disorder (ADHD). These tools will be disseminated and tested via MriCloud. We will perform three specific aims. Aim 1 will use our MriCloud architecture to deploy a Multi-Atlas Brain Mapping module for mapping an ontology of approximately 400 structures to T1 and DTI data. The architecture will support many atlases which are matched across a broad range of age from pediatric to geriatric groups, and as well as several diseases. Aim 2 will deploy a Statistical Shape Diffeomorphometry module consisting of pipelines for a) generating templates of structures from populations of cross-sectional datasets, b) data reduction to templates for cross-sectional and longitudinal geodesic mappings, and c) multiple hypothesis testing procedures based on vertex, Laplace-Beltrami basis functions and PCA basis functions. Users with their own ontology definitions of the subcortical structures will be able to generate population templates and visualize the statistics in template coordinates. Aim 3 will generate a webportal for users to use modules from Aims 1 and 2 to examine abnormalities in networks of structures such as the striatum, thalamus, amygdala and hippocampus.

Understanding how behavior emerges from the dynamic patterns of electrical and chemical activity of brain circuits is universally recognized as a fundamental mystery in science. Furthermore, better knowledge of healthy brain function has broad societal implications by laying the groundwork for advancing treatments for neurological disorders and for developing brain-inspired technologies. Various brain research efforts around the globe have embraced this scientific grand challenge. This meeting brings together US and international brain researchers to discuss the need for coordinating brain research efforts around the globe, the bottlenecks and challenges that stand in the way and strategies to overcome these. A particular focus of the meeting is on cyberinfrastructure and data resources needed to further enhance collaboration and analysis of disparate streams of neuroscience data.

The Organizers are making a strong effort to invite women and members of underrepresented groups as participants. Further, the meeting entails extensive cross-disciplinary interactions, which will be aided greatly by the face-to-face nature of this meeting. To maximize the workshop's impact, a white paper summarizing the discussion will be published at a high-profile venue.

The brain research community needs to increase the practice of sharing and combining data sets to increase the power of statistical analyses and to gain the most knowledge from collected data. This project aims to build a prototype system called BrainLab CI that will facilitate meaningful integration of thousands of publicly available Magnetic Resonance Imaging (MRI) and neurophysiology data sets, and allow researchers to define and conduct new large-scale community-level experiments on these data. BrainLab CI has the potential to transform research practice in neuroscience by overcoming major obstacles to data sharing: Scientists will be able to share data without losing control over data quality, and will maintain full visibility into how all subsequent experiments use their data and algorithms. This project may consequently drive a change in scientific culture by encouraging data sharing and the development of common analysis tools, and resulting accelerated discovery from connecting ideas, tools, data, and people. This project therefore aligns with the NSF mission to promote the progress of science and to advance the national health, prosperity and welfare.
The BrainLab CI prototype system will provide new paradigms for combining different analytic methods, meta-analysis with raw data, comparing the results of different laboratories and even synthesizing new experiments by combining different studies. An experimental-management software system will be deployed that allows users to construct community-wide experiments that implement data and metadata controls on the inclusion and exclusion of data. Example of controls include: requiring specific metadata, that data are registered to a given atlas, or that data are collected using specific experimentation protocols. BrainLab CI will initially focus on two different experimental patterns: (1) An incremental experiment defines an experiment against an existing data set which then opens to additional community contributions of data; and (2) a derived experiment forks/branches an existing experiment, allowing a researcher to change properties, such as an acceptance criteria or analysis algorithm, but otherwise run the same pipeline against the same inputs. The system will allow each experiment to maintain online dashboards showing how additional data changes results with complete provenance. To develop and validate the BrainLab CI prototype, several community experiments will be developed for MRI and for neurophysiology (including both optical and electrical physiology) data. These research domains were chosen because of the great potential gains for increased sharing of laboratory data in these domains.
This Early-concept Grants for Exploratory Research (EAGER) award by the CISE Division of Advanced Cyberinfrastructure is jointly supported by the SBE Division of Behavioral and Cognitive Sciences, with funds associated with the NSF Understanding the Brain activity including for developing national research infrastructure for neuroscience, and alignment with NSF objectives under the National Strategic Computing Initiative.

Three complimentary changes are revolutionizing the way neuroscientists study the brain. First, experimental advances allow neurobiologists to "clear" brains so that they become transparent, with the exception of a set of neurons that can be selected on the basis of their location, response properties, and genetic make-up. Second, technological advances have resulted in microscopes that can simultaneously image an entire "sheet" of this brain, thereby enabling rapid acquisition of whole brain volumes. Third, researchers are taking steps to educate neuroscientists to acquire these data. Together, this will result in a massive upswing in adoption of this experimental modality. However, acquiring the data is one step in the upward spiral of science that will yield transformative scientific results. The subsequent steps are computational. This project will develop cyberinfrastructure resources and software that enable storage and access of large CLARITY brain imaging datasets, alignment and registration to reference anatomical atlas and visualization of the datasets. Additional capabilities for automatic identification and localization of cell bodies and statistical analysis will be provided. The PIs will run annual hackathons for college students and sponsor a summer internship program for undergraduates to broaden the educational efforts in software development for neuroscience. Finally, mobile compliant digital education content will be created to complement existing online courses to target STEM students, and educate global citizens.

This project will build a prototype pipeline that operates on raw CLARITY brains and outputs the statistics of locations of cells in each region in the Allen Reference Atlas, as well as estimates of connectivity and similarity across regions and conditioned on different contexts. To do so, the PIs will leverage modern mathematical statistics (such as Large Deformation Diffeomorphic Metric Mapping for registration, Deep Learning and Random Forests for segmentation, and Statistical Graph Theory for analysis of the resulting conenctomes), as well as modern computational tools, including Docker containers to facilitate full reproducability, and semi-external memory algorithms and cloud computing to enable scalable analytics. To reach out to the broader community and educate them in the use of these tools, this project will provide tutorials deployed in the cloud. Together, this will facilitate the large community of users to both collect and analyze their data with ease. Many of the tools developed as part of this project will be easily extensible to other experimental modalities and neuroscience communities.

The purpose of this grant is to support the continued development and maintenance of DtiStudio/MriStudio/ MriCloud software family, which was developed at the PIs? lab in Johns Hopkins University. DtiStudio was introduced in 2001 and remains one of the most widely used programs to process diffusion tensor imaging (DTI) data. MriStudio, which consists of DiffeoMap and RoiEditor, was introduced in 2007 and provide a unique environment in which to perform a cutting-edge brain mapping and atlas-based image analysis. In 2014, we introduced an entirely new software platform, MriCloud, which is fully based on cloud architecture with a web- interface (www.mricloud.org). This platform not only integrated all the analysis offered by Dti/MriStudio in a fully automated manner but also provides new types of services that was made possible by the cloud architecture. Within two years, it has more than 1,200 registered users and the monthly processed data reached over 8,000 in April, 2017. There are currently four Software-as-a-Service (SaaS) provided in this platform. In this application, we propose to extend this service to the community through the following aims; Aim 1: Continued user support through update, education, and dissemination Dissemination: Currently, two major channels of dissemination are web-based resources (manuals and videos) and hands-on monthly three-day tutorials. Each tutorial accepts 12 applicants with 5 faculties and two programmers helping the attendees. As the functionalities of DtiStudio/MriStudio/MriCloud expand, the continued and most updated user supports is of the highest importance. Resource update: We continue to update them by incorporating state-of-the-art technologies and new atlas resources. The update of data I/O with ever-changing file formats (DICOM and proprietary) by the four major MRI manufactures remains essential. Aim 2: Extension of the functionality Addition of new services: We continue to work with our collaborators to incorporate further processing pipelines including structure-based lesion-load analysis using FLAIR images, MR spectroscopy, and quantitative susceptibility mapping data. QC report: We will develop and incorporate three levels of check points for quantitative data quality control. Aim 3: Integrative analysis platform: The co-existence of an array of MR image analysis pipelines within the same cloud platform provides a unique opportunity to perform multi-modal integrative analysis. We will develop tools to combine anatomical and physiological features from multiple MR contrasts and characterize unique features of a disease of interest. Aim 4: Bring to bedside: This is an exploratory aim to develop and test our cloud system as a platform for translational research.

TRACING SPREAD OF PATHOLOGY WITHIN THE HD BRAIN VIA AUTOMATED NEUROIMAGING Huntington's disease (HD) is a progressive fatal neurodegenerative disorder caused by an expanding CAG repeat in the Huntingtin gene coding for an expanding polyglutamine stretch in the Huntingtin protein. Neurodegeneration in HD is in large part caused by toxic effects of the abnormal Htt protein, and there is increasing evidence that mutant Htt can spread, like prions, and like abnormal proteins in other neurodegenerative diseases, from one neuron to another. Elucidating the sequence and pattern of atrophy in the HD brain is of special current importance, with ?gene silencing? or ?RNA-lowering? trials, using antisense oligonucleotides, or shRNA, or related reagents, in active development. The key to success of these trials will be to know where and when to intervene, since these reagents do not penetrate the blood-brain-barrier, and must be injected into the CNS. Our studies will elucidate the temporal and spatial patterns of the spread of HD neurodegeneration, to elucidate the pathogenesis of HD and to help guide interventional trials. In Specific Aim 1, we will conduct cross sectional and longitudinal analyses of the spatial and temporal pattern of volumetric change and shape change in subregions of HD compared to control brains, using longitudinal T1 and DTI scans from HD cases and controls from the PREDICT-HD study and the TRACK-HD study. Scans will undergo automated processing through MRICloud, segmented into about 400 subregions. We hypothesize that atrophy will begin in the striatum and spread sequentially to adjacent white matter and then to cortical gray matter. Alternatively degeneration may be multifocal. In Specific Aim 2 we will determine clinical correlations of the brain atrophy from Aim 1. In Specific Aim 3 we will use tract-tracing methods to study the spread of pathology in the HD brain. We hypothesize that the spread of atrophy in the HD brain follows patterns of axonal connectivity. Alternatively, it is possible that pathology begins and spreads in a multifocal fashion. Taken together these studies will delineate the longitudinal spread of pathology within the HD brain, and its clinical consequences. This information will elucidate the pathogenesis of HD and will be critical for designing the timing and localization of planned interventional trials.

SUMMARY The BRAIN Initiative seeks to accelerate the development and application of innovative technologies that ultimately will revolutionize our understanding of the human brain. A central current focus of the brain initiative is to enable a swift and comprehensive survey of all brain cell types and circuits. Recently, single cell transcriptomics is helping to classify cell types by their expression patterns. Transcriptomic analysis is currently unique in that it scales to the entire mouse brain and all cell types. However, the correspondence between transcriptomic clusters and cell types, as defined by connectivity or physiology, is not clear. Classification of cell types will require integrative analysis of at least two data elements at the cellular level: (1) molecular signature (e.g., transcriptome), (2) anatomy (e.g., location, morphology, connectivity). We therefore propose here to elevate the complete reconstruction of the morphology of molecularly defined neurons from a small-scale artisanal method to a technology that scales to the analysis of all neuron types across the entire brain thereby providing critical information about cell types, neuronal structure and connectivity. To achieve this goal, we will build on an existing research partnership between the Departments of Neuroscience and Biomedial Engineering at Johns Hopkins University and Janelia Research Campus to develop novel tools for the reconstruction of molecularly defined neuronal subtypes. We will develop scalable imaging technology and neuro-informatics tools for morpho-molecular analysis that will be made available to researchers for a community wide effort of the mapping of all neuronal cell types and connections at and unprecedented scale and depth. Preliminary data demonstrate the feasibility of our approach that will be further developed to enhance speed on a generally usable platform. We will also develop affordable tools for data storage and retrieval. Molecular identification of neurons is designed to take advantage of information from the BRAIN Initiative Cell Census Network (BICCN) to integrate data across platforms into a common Brain Atlas. Data collected in this proposal will be integrated into the searchable BICCN database at the Allen Brain Institute.

IMAGING CORE ? CORE C: PROJECT SUMMARY/ABSTRACT The BIOCARD study is a longitudinal, observational study of 349 individuals who were cognitively normal and primarily middle aged (mean age=57.1) at enrollment. The subjects have now been followed for up to 27 years. The overall objectives of the project are to further advance the study of preclinical Alzheimer?s disease by: (1) clarifying the pattern and rate of change in AD biomarkers (including those based on CSF, blood, MRI, and PET imaging) and cognition; the biomarkers to be studied include several promising novel biomarkers derived from blood, CSF and brain imaging. (2) maximizing our data by working collaboratively with several research groups who have comparable data, and (3) providing a publicly accessible data, brain scans, and biological specimens, for researchers in the field. To accomplish these goals we established 7 Cores and, with this application, are also including 2 projects. The Imaging Core (Core C) is responsible for overseeing the acquisition and analysis of magnetic resonance imaging (MRI) scans and positron emission tomography (PET) scans in the BIOCARD study participants. The specific aims include: (1) to facilitate the continued acquisition of 3T MRI scans in the participants, (2) to facilitate the continued longitudinal acquisition of PET scans, using Pittsburgh Compound B (PiB), and the initiation of Tau PET imaging, using MK6240, (3) to integrate the 1.5T volumetric data collected at the NIH with the 3T MRI volumetric data collected at Johns Hopkins, (4) to integrate the MRI and PET imaging data with the clinical, cognitive, CSF and blood biomarker data generated by Cores B and D and Project 1, (5) to examine the order and timing of MRI and PET biomarker changes during preclinical Alzheimer?s disease, (6) to integrate high field MRI images of brain tissue with detailed neuropathological findings in the same individuals, and (7) to continue to share image files and related clinical information with investigators in the field.