Richard M. Leahy - US grants

9422106 Leahy This award is to purchase equipment dedicated to research in computer and information science and engineering. Specifically, the equipment will be used for research in multi-dimensional signal and image processing, including in particular: 1) fusion of multimodal neuroimaging data; 2) adaptive quantization of image and video; 3) automatic target recognition via deformable template matching; 4) design of high resolution diffractive optics for photonic interconnections and computing; and 5) advanced adaptive multidimensional and array signal processing. Common to all of these projects is a need for access to fast numerical computation and high resolution visualization and display capabilities. The goal of this project is to set up a state of the art facility for processing, visualization and display of multidimensional data. Towards this end, a computer for high performance numerical computation, and a RAM-based workstation for display of high resolution video image sequences with a high performance graphics capability will be purchased. ***

The magnetoencephalogram (MEG) and electroencephalogram (EEG) provide unique insights into the dynamic behavior of the human brain as they are able to follow changes in neural activity on a millisecond timescale. In comparison, the other functional imaging modalities (positron tomography (PET) and functional magnetic resonance imaging (MRI) are limited in temporal resolution to time scales on the order of, at best, one second by physiological and signal-to-noise considerations. However, the importance of EEG and MEG as functional imaging modalities has been diminished by the absence of definitive studies demonstrating their ability to provide accurate spatial localization of complex sources either in vivo or in a realistic head model. The absence of such studies is due to the highly ill-posed nature of the inverse problem, and to a concentration in the EEG/MEG community on human and animal experiments where true validation is often difficult. Another factor limiting progress towards the use of EEG/MEG in studying brain activation has been the lack of inverse algorithms capable of fully utilizing the spatial and temporal information collected during sensory, motor or cognitive activation. Furthermore, recent advances in the other anatomical and functional imaging modalities offer the potential for bringing a wealth of additional information to the problem. There is therefore a clear need both for the development of new algorithms which exploit the most recent advances in sensor design, signal processing theory, and other functional and anatomical imaging modalities, and a detailed study of the limitations of these and existing inverse procedures. It is the goal of this project to develop new algorithms to exploit the full potential of EEG and MEG based source estimation. Models based on multiple dipoles and distributed current source will be developed. In both cases we will consider parametric and nonparametric models for the associated temporal activity. Spatial constraints on the source locations, based on volume MRI and functional studies (fMRI or PET), will be incorporated in these models where appropriate. The forward model will allow incorporation of realistic head geometries as well as locally fitted spheres. Starting from a general Bayesian framework incorporating all of the above, we will develop inverse procedures for a number of specific spatio-temporal source configurations. All procedures will be studied using (i) theoretical tools such as the Cramer-Rao lower bound; (ii) Monte-Carlo studies of bias, variance and robustness to modeling errors; and (iii) experimental evaluation using special phantoms including one constructed from a human skull containing multiple dipole and distributed current sources. Platform independent software and documentation for these algorithms will be distributed to interested researchers in the brain imaging community. In addition to providing a suite of thoroughly tested inverse procedures, we anticipate that this work will provide important insights into the fundamental limitations of EEG and MEG based source estimation.

DESCRIPTION (Adapted from Applicant's Abstract): The new generation of PET scanners is approaching the resolution limit of the modality through the use of new scintillators, smaller detectors and 3D data acquisition. However the filtered backprojection (FBP) image reconstruction schemes used in most clinical settings are unable to fully realize the potential of these systems as they do not take account of the photon-limited nature of the data or the fact that the coincidence data do not represent perfect collimation between individual detector pairs. The thesis of this project is that PET image reconstruction is best performed within a statistical framework in which the physical and statistical properties of the observed data are accurately modeled. In the current funding period fast 2D and 3D Bayesian reconstruction methods were developed which combine accurate statistical and physical models of coincidence detection with image models designed to reflect the piece-wise smooth nature of in-vivo tracer distributions. These methods have been compared to standard clinical protocols and shown to produce improvements in resolution at matched background noise levels, quantitative accuracy, contrast recovery, and lesion detection using observers. During the proposed project period they will continue to extend the Bayesian reconstruction methods to a range of state of the art scanners: (I) whole body CTI EXACT HR+, (ii) the new CTI high resolution brain scanner with LSO based depth-of-interaction detectors, (iii) the simultaneous x-ray CT/PET scanner being developed jointly by CTI and the University of Pittsburgh, (iv) the UCLA microPET small animal scanner, and (v) the dedicated LSO-based PET mammography system under development at UCLA. They will use the factored matrix method to model positron range and photon-pair angular separation, geometric and intrinsic detector efficiencies, crystal penetration and scatter, and dead-time. Within the conjugate gradient Bayesian image reconstruction framework they will include a range of Gibbs priors (e.g. convex pair-wise smoothing, anatomically-based priors, spatially variant weightings for uniform resolution) and likelihood functions (e.g. Poisson, Gaussian approximations, modified Poisson models to account for the effects of randoms subtraction). They will include Ordered Subsets EM (OSEM) implementation options in the code for comparative purposes and to allow fast reconstruction. They will also implement a list mode likelihood variation of our methods. Optimized multithreaded implementations of these image reconstruction methods will be developed for use on single and multiprocessor UNIX workstations and Pentium-Pro servers. Performance of Bayesian, OSEM and optimized FBP reconstruction methods will be evaluated through studies of resolution, signal to noise measures, quantitation and lesion detection. Quantitation studies over small and irregularly shaped regions of interest that include consideration of partial volume effects will be performed using phantoms and combined autoradiography, MR and PET studies of small animals. Lesion detection will be studied using data from a realistic breast & thorax phantom and simulated lesions added to human data. Our optimized software will be made available to our collaborators and other interested research sites.

The magnetoencephalogram (MEG) and electroencephalogram (EEG) provide unique insights into the dynamic behavior of the human brain as they are able to follow changes in neural activity on a millisecond timescale. In comparison, the other functional imaging modalities (positron tomography (PET) and functional magnetic resonance imaging (fMRI)) are limited in temporal resolution to time scales on the order of, at best, one second by physiological and signal-to-noise consideration. The goal of this project is to develop and evaluate computational techniques for estimating the location, extent and dynamic behavior of the current sources that produce the observed MEG and EEG. During the initial project period, we have developed a suite of methods and software for head modeling, source localization and imaging. We have also built and tested a skull based phantom and developed computational tools for comparing and quantifying performance of different models and inverse methods. We plan to build on this work in the proposed project period, by concentrating on using and extending the methods we have developed to date to address several fundamental questions of relevance to both EEG/MEG researchers and the brain imaging community as a whole: (i) How reliably can E/MEG find the locations of multiple current sources in the brain? (ii) To what extent can E/MEG determine the spatial extent of distributed current sources? (iii) How accurately can we find the time series or activation sequence of these sources? (iv) How do we best process data from cognitive studies involving the differences between conditions? (v) How is E/MEG data best combined with functional MR or PET activation data? Extensions of the methods that have been developed during the initial project period will include techniques for relating dipolar and multipolar estimates to neural activity in the cerebral cortex and methods for combining these estimates with fMRI data. We will also develop a source localization methodology for processing cognitive data which allows identification and removal of the signal components common to two different test conditions. The Bayesian imaging method developed during the initial project period will be extended to utilize fMRI data in the prior. For our multipolar and imaging methods, we will also examine the impact of different head models on computation cost and accuracy. Performance of all methods will be evaluated using a range of computational, phantom and human data. Computational tools include Cramer-Rao lower bounds, Monte-Carlo methods and the use of subspace correlations. Phantom data will be generated using the realistic 32-dipole human skull phantom that was constructed during the initial project period. Human data will be based on simple motor paradigms well documented in the literature, for which MEG, EEG and fMRI data will be collected. Software and data from this project will be made available to researchers via the Internet.

DESCRIPTION (Provided by Applicant): The purpose of this project is to develop, implement, analyze and evaluate methodologies for estimating the location, extent, and dynamic behavior of cortical neuronal populations that give rise to event-related electroencephalographic (EEG) and magnetoencephalographic (MEG) (E/MEG) signals. During the lifetime of this project we have developed Bayesian cortical imaging methods and multipolar source localization methods, both of which are able to localize focal neuronal populations in cortex. We have also developed fast, accurate forward models based on boundary element methods (BEMs). These developments are embodied in an interactive software package, BrainStorm, which is now available to the research community. Evaluation of our methods and software has been based on theoretical and Monte Carlo studies, studies using a human-skull phantom developed under this project, and applications to event-related EIMEG data. A high-resolution cortical surface, and skull, scalp, and brain surfaces for BEM calculations, are found using a series of automated processing steps applied to anatomical MRIs which are embodied in a second software package BrainSuite. For the proposed project period we plan to continue our investigations of the theoretical basis of E/MEG source estimation with the goal of better understanding the potentials and limitations of the modality. Our theoretical investigations will lead to improvements in forward and inverse methods and, importantly, a better understanding of the uncertainties implicit in these estimates. In verse methods will be based on the use of regularized signal-subspace localization of multipolar sources that can represent distributed neuronal populations. Cortical images will be obtained by re-mapping multipolar sources to cortex. We will develop tools for estimating location uncertainty directly from the measured data and estimated sources. These methods will be based on the bootstrap method in statistics and use of plug-in approximations to the Cramer Rao lower bounds. The cortical remapping methods will optionally allow the use of fMRI activation sites as prior information on possible source locations. A novel method will be developed for selection of event-related components in raw data for use in cases where stimulus-locked averaging obscures sources with highly variable latency. Improved forward models will be based on finite element methods (FEMs) applied to accurately segmented MRIs of the subjects head. For subjects for whom MRIs are not available, we will develop a generic head model based on BEM or FEM applied to an averaged head warped to a series of digitized scalp landmarks. Evaluation will be based on continued theoretical, simulation, and phantom studies. These technical developments will be incorporated in our BrainStorm and BrainSuite software packages which we will continue to maintain and enhance over the life of the project. Limited human studies of motor function using fMRI, MEG and EEG are planned for this project; broader applications of these methods in functional mapping and epilepsy will be realized through use of our software by our collaborators and other registered users.

[unreadable] DESCRIPTION (provided by applicant): [unreadable] The "International Symposium on Biomedical Imaging: from Nano to Macro" is a forum for researchers involved in the following aspects of biomedical imaging: physical, biological and statistical modeling, image formation and reconstruction, computational image analysis, statistical image analysis, visualization, and image quality assessment. The meeting aims to facilitate cross-fertilization between different imaging modalities and an integrated approach to biomedical imaging across scales. We encourage submission from researchers involved in applications at the nano, molecular and cellular levels through macroscopic and whole-body clinical systems. Modalities of interest include gene expression microarrays, electron, confocal and multiphoton microscopy, optical and fluorescent imaging and spectroscopy systems, autoradiography and in situ hybridiziation, x-ray and nuclear tomography, anatomical and functional MRI, ultrasound, magneto and electroencephalography, and small animal imaging systems. Imaging applications of interest include gene expression mapping, drug discovery and delivery, molecular imaging, functional brain mapping, computational neuroanatomy, cardiac imaging, and cancer imaging. The meeting is the second in a series co-sponsored by two IEEE societies: the Signal Processing Society (SPS) and the Engineering in Medicine and Biology Society (EMBS). The inaugural ISBI meeting was held in July 2002. With over 500 attendees, this very successful meeting featured opening addresses from Dr. Elias Zerhouni, Director of NIH, and Roderic I. Pettigrew, Director of NIBIB. The success of the first ISBI meeting in 2002 bodes well for its future. However, the inherently interdisciplinary nature of the field means that no single professional organization has the majority of potential participants as its members. As a new meeting ISBI is more likely to attract members of IEEE than non-society members. We anticipate that after the first few meetings, ISBI will be sufficiently established that non-IEEE members will plan to attend because of the quality of the meeting. However, for the meeting planned for 2004 we believe it is essential to reach out to other communities, not only through publicity, but also through targeted invitations to organize and present special sessions dealing with currently important research issues in biomedical imaging. This proposal is submitted for the purpose of providing funds to defer the costs of attendance of selected invited speakers in these sessions. We will also encourage participation by new investigators by providing travel stipends to qualified participants (postdoctoral fellows and junior faculty) to present contributed papers. Finally we will plan to provide a number of passes for students and postdocs to attend the short courses on the first morning of the meeting. [unreadable] [unreadable]

Abstract

Program: NSF 04-588 CISE Computing Research Infrastructure
Title: CRI: Reconfigurable computing infrastructure for high end and embedded computing applications
Proposal: CNS 0454407
PI: Prasanna, Viktor K.
Institution: University of Southern California

The investigators will acquire a reconfigurable computer comprised of general purpose processors, field programmable gate arrays (FPGAs), a common memory, and an interconnect fabric joined under a programming model that works with all the parts. The acquisition of this machine will enable research at a realistic scale on actual reconfigurable machines for performance testing, validation, and applications demonstrations. This infrastructure will be robust enough to implement application "kernels" such as (e,g, an LU implementation or n-body simulation) that give realistic scale experimental results. Applications that will be explored include matrix operations, computational genomics, molecular dynamics, density functional theory, and finite element methods. The team will also be able to work on energy efficiency for embedded FPGAs. Broader impacts of this project include the potential impact on reconfigurable systems, use of FPGAs for applications, and discoveries in the applications areas. The investigators participate in USC's Minority Opportunities in Research (MORE) program.

This project is focused on the development, implementation, validation and distribution of computational tools for the analysis of MEG and EEC (E/MEG) data for the purpose of developing an increased understand- ing of normal and pathological function in the human brain. Over the past four years we have made progress in the following areas: (i) the development of tools for the analysis of anatomical MR data for use in localizing E/MEG sources; (ii)computationally efficient and atlas-based forward models; (iii)parametric and image- based inverse MEG and EEG methods; (iv) statistical analysis of parametric and image-based inverse solu- tions; (v) software development, distribution and support; (vi) validation with both simulated and in vivo data; and (vi) applications to both epilepsy and cognitive neuroscientific data. This work focused primarily on analy- sis of event-related averaged data in single subjects. Statistical analysis was restricted to pair-wise compari- sons. For the next project period our goal is to extend our inverse approaches to investigate both evoked and induced responses using imaging and parametric methods and to extend our statistical tools to allow analysis of experimental data involving multiple conditions in individuals and groups. We will also develop tools to investigate interactions between brain regions and again develop statistical tools to assess the significance of these models. These methods will be validated using computer simulation and in vivo studies involving visu- ally cued attention paradigms in normal subjects. The specific aims that we will address in developing and evaluating these tools are as follows: (1) Inverse Methods: development of linear imaging methods optimized for maximally isotropic and shift invariant resolution, cortically constrained dipole and multipole fits for estima- tion of multiple focal sources, and beamforming methods for simultaneous monitoring of multiple distinct corti- cal regions; (2) Cortical Alignment: development of tools for coregistering cortical surfaces using covariant PDEs based on alignment of sulcal features and minimization of a thin plate spline bending energy in the intrinsic geometry of the cortical surface; (3) Statistical Analysis: development of nonparametric methods for univariate analysis of cortical maps of induced and evoked responses in individuals and groups; (4) develop- ment of models of large scale cortical interactions using broadband linear and nonlinear methods based, respectively, on the multichannel autoregressive model and bispectral analysis; (5) distribution of software implementations of the methods in Aims 1-4through continued development of the Brainstorm and BrainSuite software tools; and (6) application of all of the above methods to the analysis of experimental high density EEG and MEG data from a visually cued attention study to investigate the ability of the methods developed in Aims 1-4 to extract a meaningful and consistent picture of brain activation and communication.

The brain-basis of perception is complex, and recent research suggests that neural processing depends on large-scale oscillation of neuronal units. Oscillatory cortical networks detected with electroencephalography and magnetoencephalography recordings often involve several frequency bands, indicating that a multivariate (multi-frequency) analytic approach would have better sensitivity in detecting neural effects than univariate analysis. However, popular connectivity measures, such as coherence and phase synchrony, typically analyze pairs of spatial locations and take into account a single quantity from each location, such as amplitude or phase within a specified frequency band. With funding from the National Science Foundation, Drs. Dimitrios Pantazis, Richard Leahy, and Jintao Jiang will develop robust multivariate statistical methods for detecting brain interactions in electroencephalography. Given the wealth of information in electroencephalography data, analysis using a single frequency approach requires either prior knowledge of the frequencies at which interactions occur or, conversely, a large number of tests, one for each possible type of interaction. In this project, the researchers are using canonical correlation analysis, which can find the optimal combinations of frequencies in one cortical site that best correlate with frequencies at another cortical site. In contrast to conventional methods of interaction analysis, this project is automating the identification of frequency bands that contribute significantly to cortical networks. The target application focuses on audiovisual speech integration effects. The multivariate methods developed in this proposal are being used to detect multisensory interaction cortical sites and account for different levels of phase-resetting from audiovisual speech stimuli with different stimulus onset asynchronies.

This research will facilitate the detection of oscillatory cortical networks both in the normal and pathologic brain. Changes in oscillatory brain activity have been reported in a wide array of neurological diseases, including epilepsy, schizophrenia, and Alzheimer's disease, and improved methodologies to detect the presence and differences in oscillatory activity and associated networks will in turn advance the understanding of these diseases and facilitate the development and assessment of therapeutic interventions. This effort brings together engineers and neuroscientists to tackle a broad range of scientific and technological problems, and as a result, the project offers opportunities for integrated interdisciplinary research training of doctoral students. Research results will be disseminated broadly to the research community through professional meetings and journals, and the developed research tools will be distributed to the research community through the open source software BrainStorm.

DESCRIPTION (provided by applicant): Fully 3D time-of-flight (TOF) PET scanners offer the potential for previously unachievable signal to noise ratios in clinical PET. Relatively new, fast scintillators have the combination of high speed, stopping power and light output, that make clinical TOF PET practical. Consequently it is likely that TOF will become the standard for clinical whole body PET in the near future. We will build on our experience in PET image reconstruction methodology and our recent results on rebinning of time of flight data to develop image reconstruction methods that are optimized for use with TOF-PET data. In non-TOF PET systems we have seen a progression over the past decade in the methods used for clinical studies from analytic reconstruction, to those based on Fourier rebinning and 2D iterative reconstruction, to fully 3D iterative reconstruction. The reason for this is that iterative 3D reconstruction using all of the data and more accurate models can achieve improved performance relative to the other approaches. Similarly, using the full TOF data in an iterative reconstruction framework should also lead to the best performance. The main goals of this project are: (i) to develop an optimized fully 3D TOF reconstruction method that extends our earlier MAP approach for 3D PET to TOF; and (ii) to systematically study the trade-offs involved in developing faster TOF PET reconstruction algorithms and to implement and evaluate practical methods with a computational cost consistent with their use in clinical settings. We will first develop a TOF-extension of our MAP approach to reconstruction that combines accurate physical and statistical modeling with fast convergent algorithms. Spatially variant penalty functions will be used to ensure count independent and spatially invariant resolution. We will then use this as a benchmark against which to compare other simplified methods. Among these we will investigate the use of novel Fourier rebinning methods for fast forward and backprojection of TOF-PET data, as well as data reduction methods in which we use rebinning to reduce TOF to non-TOF data. These investigations will include a study of the mathematical properties of the TOF-PET data, analysis of the bias and covariance of MAP reconstructions, and the development of fast computational algorithms. The result of this project will be a combination of practical software for TOF- PET image reconstruction with analytic studies of the properties of these methods and computational, phantom and human-study evaluation of these reconstruction methods.

DESCRIPTION (provided by applicant): Magnetoencephalography (MEG) and electroencephalography (EEG) provide a unique window to the large scale spatiotemporal neural processes that underlie human brain function. However, even with restrictive models, the low SNR, ill-posedness of the inverse problem, and difficulty of differentiating ongoing brain activity and other electrophysiological signals from induced and event-related changes result in unique challenges in data analysis and interpretation. These challenges include accounting for artifacts and noise, accurately forward modeling from cerebral sources to sensor space, defining appropriate source models, computing inverse solutions, and detecting and quantifying interactions. In this grant we will continue the development of our linked MNE-Python and Brainstorm software packages. Emphasis in the current software is on data preprocessing, the formation and statistical analysis of inverse solutions, and advanced, interactive display and interpretation of these solutions. We have established standard workflows for cortical current density mapping, time-frequency analysis, and statistical testing in both MNE and Brainstorm for these procedures. In the next project period in Aim 1 we will build on these procedures adding new dimensions to the data workflows for the interaction measures described in Aim 3. Under this aim we will also continue general software development (including automated testing and documentation), support and dissemination activities for users. In Aim 2, we will expand the use of Python-based scripting to facilitate large-scale batch processing of multiple subjects and/or conditions from an extensive experimental study. We will also add the ability to import locations of intracranial EEG sensors (depth electrodes and cortical grids) for display and interaction analysis using methods from Aim 3 and for cross-validation of MEG/EEG non-invasive source models. To fully realize the potential of EEG/MEG to elucidate the spatio-temporal networks that underlie human perception, cognition, and action, we will also develop tools to investigate the interactions between cortical neuronal populations. These tools should take into account the dynamically nature of these networks, the inherent complexity of causal and inter-frequency interactions amongst neural populations, and the fact that interactions can occur between multiple brain regions. Since no single parsimonious model can account for all such interactions, Aim 3 of this grant will develop a suite of interaction modeling and powerful visualization tools for use by neuroscience and clinical researchers.

? DESCRIPTION (provided by applicant): Treatment of intractable focal epilepsy by resection of the seizure onset zone (SOZ) is often effective provided the SOZ can be reliably identified. Focal epilepsy, however, is fundamentally a network-based disease. The seizure onset zone is connected to a network whose other nodes may also exhibit abnormal neural activity either concurrently or subsequently. In patients without MRI detectable lesions, differentiation of the onset zone from these other nodes in the network can be difficult, even with the use of invasive recordings. The goal of this project is to improve SOZ identification, ultimately reducing the need for presurgical invasive recordings where possible, and guiding placement of electrodes in those patients who do need invasive monitoring. To achieve this goal, in Aim 1 we will build a functional connectivity atlas from a database of invasive Cortico-Cortical Evoked Potential (CCEP) recordings to identify common interaction networks in patients with partial epilepsy and to investigate the degree to which these are dependent on the location of the SOZ. To construct the atlas, patient data will be coregistered to a labelled anatomical atlas using a cortically constrained warping of each subject's structural MRI. In Aim 2, CCEPs data will be supplemented in the atlas with other data that provide additional insight into the brain regions involved in the seizure: regions of hypometabolism in interictal FDG PET, hypermetabolism in ictal SPECT, interictal spike localization from EEG and MEG and invasive recordings, functional areas associated with seizure semiology, MR-identified lesions, area of resection, post-surgical Engel classification. Using machine-learning methods, we will perform a sequence of tests to examine the degree to which the atlas can be used to identify the SOZ in individual subjects. Finally, in Aim 3, we will investigate the potential for using regional connectivity established frm noninvasive MEG data and resting state MRI in combination with the CCEPs atlas to identify these networks, with the ultimate goal of reducing the need for invasive monitoring. Retrospective analysis using a leave-one- out approach and comparison with outcomes will be used to quantify improvement in identification of the onset zone from both invasive and noninvasive recordings.

Abstract For the past 20 years, we have developed and distributed BrainSuite, a collection of open-source software tools that provide advanced capabilities for analysis and visualization of brain MRI. Since 2011, this effort has been largely supported by the parent NINDS Grant R01-NS078940. The parent grant was awarded under the ?Extended Development, Hardening and Dissemination of Technologies in Biomedical Computing, Informatics and Big Data Science? mechanism (PA-14-156), and is largely focused on developing new tools for coregistration of multimodal imaging data, modeling and analysis of diffusion data, and quantitative analysis of functional and structural connectivity. These tools have been developed with the goal of supporting a broad a range of neuroimaging studies, but we have not previously had an Alzheimer?s disease (AD) on this project. In the present application, we identified key areas where BrainSuite can be applied and extended to create new methods specifically designed for probing brain imaging data to identify functional and structural changes that indicate neurodegeneration in subjects in the early stages of Alzheimer?s disease. These new methods will focus on two areas of the brain, the transentorhinal and entorhinal region of the medial temporal lobe, that are known to show tau tangles in the earliest stages of the disease. The transentorhinal cortex in particular is one that has not often been analyzed using computational anatomy methods despite its implication in early-stage AD. Our proposal builds on recently developed longitudinal diffeomorphic analysis methods and applying them to the study of the transentorhinal cortex (TEC) and entorhinal cortex (ERC), which relied upon manual delineation to identify TEC and ERC. Given the high potential for these new methods to be of value in AD research, we have developed Aim 1 of the supplement proposal to adapt BrainSuite?s registration and labeling tools, developed under the parent grant, to automatically identify TEC and ERC. This will accelerate the use of the longitudinal diffeomorphic analysis methods. Under Aim 2 of the supplement proposal, we will develop the longitudinal diffeomorphic pipeline as a tool integrated into BrainSuite. This will accelerate the ability of investigators to apply this framework in the research setting. In Aim 2, we will also develop tools to examine AD-related functional changes in resting-state fMRI data. This will build on the methods we have developed in the parent grant for analyzing fMRI data, and enable the examination of changes in functional connectivity associated with TEC and ERC that are likely to result as a product of neurodegeneration associated with AD. We anticipate that the new research supported by this supplement will lead to new tools that will provide an AD biomarker that provides an early indicator of AD, as well as a mechanism for tracking AD progression. As with all BrainSuite tools, these new methods will be made available freely as open source software. In this way, we expect that these tools will lead to new studies, both by other investigators as well as our own team, that use these tools in imaging studies of AD.

Project Summary Electrophysiological recordings in humans and animals play an essential role in developing an understanding of the human brain. Signal recording technology spans the entire scale from invasive microelectrode single-unit recordings, through mesoscale macroelectrode measures of local field potentials, to whole-brain monitoring through measurement of scalp potentials (EEG) and extracranial magnetic fields (MEG). Analysis of these data presents a host of challenges, from low level noise removal and artifact rejection to sophisticated spatio-temporal modeling and statistical inference. The multidisciplinary neuroscience research community has an ongoing need for validated and documented open-source software to perform this analysis and to facilitate reproducible and large-scale research involving electrophysiological data. This proposal describes our plans to continue to develop and support Brainstorm, open-source software that meets this need. Brainstorm is a Matlab/Java multi-platform (Linux, MacOS, Windows) software package for analysis and visualization of electrophysiological data. The software is extensively documented through a series of detailed tutorials and actively supported through a user forum and a mailing list. Over the past 8 years we have registered 16,000 distinct users, provided hands on instruction to 1,200 trainees, and the software has been used and cited in ~600 journal papers. Brainstorm includes tools for importing MEG/EEG, intracranial EEG, animal electrophysiology, and near-infrared spectroscopy (NIRS) data from multiple vendors, extensive interactive features for data preprocessing, selection and visualization, coregistration to volume and surface MRIs and atlases, forward and inverse mapping of cortical current density, time-series and connectivity analysis, and a range of statistical tools. Data can be analyzed through a graphical interface or through scripted pipelines. The current proposal represents a plan to extend Brainstorm in a manner that leverages the unique features of our software and addresses important needs for large-scale data analysis. In this project we will continue to extend and support our software through the following three specific aims: (i) we will harness recent developments in distributed and shared data and high performance computing resources, together with standardization of data organization, to facilitate large-scale, reproducible analysis of electrophysiological data. (ii) We will also address the need for improved modeling resulting from the increasing use of both invasive recordings and direct brain stimulation through development of new modeling software for accurate computation of the intracranial electromagnetic fields produced by brain stimulation and neuronal activation. (iii) Finally, we will continue to add new functionality and to support the software through in-person training, online forums, documentation and other resources.