Brain Mapping seminar abstracts 2011-12

Gael Varoquaux, PhD
INSERM, Post-doctoral fellow, Unicog team, NeuroSpin, supervision: A. Kleinschmidt INRIA, Parietal team, NeuroSpin, supervision: B. Thirion, J.P. Poline
Research interests: computational and statistical modeling of complex systems. My current research is on building quantitative probabilistic and descriptive models of brain function through computational analysis of functional imaging. I am also interested in the development of scientific computing tools and software engineering best practices for computational sciences. My original training is in physics. I did my PhD on quantum-based atom-optic metrology.

Learning and comparing multi-subject models of brain functional connectivity
High-level brain function arises through functional interactions. These can be mapped via co-fluctuations in activity observed in functional imaging. First, I will show how spatial maps characteristic of on-going activity in a population of subjects can be learned using multi-subject decomposition models extending the popular Independent Component Analysis. These methods single out spatial atoms of brain activity: functional networks or brain regions. With a probabilistic model of inter-subject variability, they open the door to building data-driven atlases of on-going activity. Subsequently, I will discuss graphical modeling of the interactions between brain regions. To learn highly-resolved large scale individual graphical models models, we use sparsity-inducing penalizations introducing a population prior that mitigates the data scarcity at the subject-level. The corresponding graphs capture better the community structure of brain activity than single-subject models or group averages. Finally, I will address the detection of connectivity differences between subjects. Explicit group variability models of the covariance structure can be used to build optimal edge-level test statistics. On stroke patients resting-state data, these models detect patient-specific functional connectivity perturbations.

Dost Ongur, MD, PhD
Clinical Director Psychotic Disorders Division, McLean Hospital Assistant Professor in Psychiatry, Harvard Medical School
I have an M.S. degree in Neuroscience from Yale University, after which I joined the Medical Scientist Training Program at Washington University in St.Louis where I obtained my M.D./Ph.D. degrees. I obtained psychiatric residency training at the MGH/McLean Adult Psychiatry program (2000-2004) involving clinical care in emergency room, inpatient and outpatient settings. Since graduation from residency, I have been working on my research with the goal of developing an independent federally-funded research program focusing on neuroimaging in schizophrenia and bipolar disorder.

Novel MRS approaches to probe white matter abnormalities in schizophrenia
White matter abnormalities are widely reported in schizophrenia and contribute to the characterization of this disorder as a "disconnection syndrome". Much work to date has used Diffusion tensor imaging, but this approach does not provide specific information about biological abnormalities underlying the detected abnormal signal. We are using a complementary approach with NAA diffusion to specifically examine intra-axonal abnormalities in the PFC white matter in schizophrenia at 4 Tesla. We also collect water and metabolite T2 relaxation times as well as magnetization transfer in the same subjects. As will be discussed, we find evidence for both abnormal NAA diffusion and reduced myelination in our data.

Eric Klawiter, MD, MSc
Assistant Professor of Neurology at Harvard Medical School Department of Neurology, MGH
Eric Klawiter is an Assistant Neurologist at Massachusetts General Hospital who specializes in multiple sclerosis (MS) and related diseases. He received his medical training at the Sanford School of Medicine at the University of South Dakota. Following medical school, he completed a medical internship and residency in Neurology at Barnes-Jewish Hospital and Washington University in St. Louis. Dr. Klawiter received fellowship training in neuroimmunology (multiple sclerosis) at Washington University where he was the recipient of a Clinical Research Training Fellowship from the American Academy of Neurology. Dr. Klawiter's research interests include multiple sclerosis clinical research and developing new imaging techniques to better understand, diagnose and treat multiple sclerosis. He has written numerous publications and book chapters relating to multiple sclerosis and neuromyelitis optica. Additionally, he has participated in multiple clinical trials of multiple sclerosis therapeutic agents.

Imaging Correlates of Cognitive Dysfunction in MS
Cognitive dysfunction is common in MS and contributes to loss of employment and reduced quality of life. This talk explores imaging correlates of cognitive dysfunction in MS. By exploring connectivity of brain regions, resting-state functional connectivity MRI has the potential to improve the understanding of MS pathogenesis while serving as a unique means of examining cognition. Focal demyelination has a predilection for the corpus callosum in MS and callosal atrophy is a common finding. We examine inter-hemispheric connectivity to evaluate the functional role of the corpus callosum in cognitive dysfunction in MS. Additionally, we will explore how cortical pathology potentially affects connectivity in a locally constrained area, termed local connectivity.

Mark Slifstein, PhD
Associate Professor of Clinical Neurobiology Dept. of Psychiatry, Columbia University
Dr. Slifstein is the director of the Brain Imaging Core of the Silvio Conte Center for Schizophrenia Research and the director of image analysis in the Division of Translational Imaging in the Dept. of Psychiatry at Columbia University and the New York State Psychiatric Institute. He received his doctorate in mathematics from New York University in 1999 at which time he began his career in PET image analysis at Columbia University. His research interests include pharmacokinetic modeling methodology and all aspects of quantitative image analysis. He is a member of the editorial board of Neuropsychopharmacology and has published over 50 peer reviewed articles on PET imaging.

The Occupancy Study in PET imaging: From the simple to the not-so-simple
One of the most robust applications of neuroimaging with PET is the occupancy study, in which the fraction of neuroreceptors occupied by an exogenously administered drug is inferred through competition with a radiotracer. The fundamental model is straightforward, but dependent on several basic assumptions, for example, that the drug and the tracer are both highly selective for the same target, or that the drug concentration is in steady state during the measurement. Sometimes the system under study meets all of these criteria very well but frequently it does not. In this talk, we will look at the methodology of occupancy studies, from basic to complex situations, through examples, analysis and simulation. The intention, in addition to highlighting some specific studies, is to convey the general scope and flavor of competitive binding models applied to in vivo molecular imaging.

Peter Caravan, PhD
Assistant Professor in Radiology at Harvard Medical School Assistant in Chemistry at Massachusetts General Hospital
Peter Caravan received his B.Sc. (Honors) at Acadia University and his Ph.D. in chemistry from the University of British Columbia (under the guidance of Chris Orvig), followed by an NSERC postdoctoral fellowship at the Universite de Lausanne (under the guidance of Andre Merbach). He then spent 9 years at Epix Pharmaceuticals developing tissue-specific and responsive MRI contrast agents where he was ultimately responsible for all chemistry and contrast agent research at the company. He is co-inventor of EP-2104R, a fibrin-specific contrast agent for thrombus detection, that is the first molecular MR imaging agent to enter into human clinical trials. He joined the Radiology faculty at Harvard Medical School and Massachusetts General Hospital in 2007. Dr. Caravan's current research focus is on the development and application of new PET and MR imaging probes and in exploiting the inherent synergies in dual PET-MR imaging.

Thrombosis, Fibrosis, and the B0 Shuffle
This seminar will summarize our recent efforts in molecular imaging of thrombosis and fibrosis, as well as our attempts to make molecular imaging more specific and quantitative. The seminar will be in four parts. I will first describe our efforts to develop a thrombus specific PET probe. Identification of thrombus remains a major unmet need in many pathologies, e.g. finding the source of embolic stroke. Presence or absence of thrombus has a major impact on patient management, and an accurate imaging test would guide clinical decisions. We are developing PET probes based on fibrin-specific peptides labeled with Cu-64 or F-18 isotopes to address this need. I will then discuss our studies aimed at developing a noninvasive means to indentify and stage fibrosis. Here we are using a type I collagen targeted MR probe. We have used this probe to identify liver fibrosis in rodent models and to correlate imaging biomarkers with total collagen and with histological scoring of fibrosis. Fibrosis is a common result in many chronic diseases and we have begun to look beyond the liver to address fibrosis imaging in the heart, the lungs, and in fibrotic tumor models. Finally, I will discuss our efforts to make molecular imaging more specific and quantitative. I will describe the 'delta relaxation enhanced MR' or dreMR technique. This involves switching the B0 field by as much as Ī0.25T about the main field. I will show that this results in contrast that is specific to slow tumbling paramagnetic contrast agents, e.g. protein bound probes. This method has the potential to quantitate probe and target protein, and results in direct protein imaging. A second approach is the development of responsive bimodal MR-PET probes. It has long been established that certain MR probes can have relaxivity which changes with external stimuli. However the MR signal change post contrast agent is dependent on both relaxivity of the contrast agent and its concentration. By incorporating a PET reporter we show that it is feasible to use simultaneous MR-PET to directly quantify the probe, lifting this ambiguity.

Jacob M Hooker, PhD
Assistant Professor at Harvard Medical School Assistant Chemist, Assoc. Director of PET Core at Masssachusetts General Hospital Director of Radiochemistry at Athinoula A. Martinos Center
Assistant Professor in Radiology at Harvard Medical School Assistant in Chemistry at Massachusetts General Hospital Department of Radiology, MGH

Visualization of Chromatin-modifying Enzymes with Positron Emission Tomography
A number of enzyme-catalyzed processes, which modify the DNA molecule and its associated chromatin, have been identified. These epigenetic processes can alter gene expression and result in major changes to cell function. Not surprisingly, dysfunction and dysregulation of these processes have been associated with human disease. In the brain, the consequences of epigenetic changes (results of gene-environment interactions) can be extraordinary—leading to impaired cognitive function. In some cases, these associations have led to new therapeutics opportunities that "rescue" brain function by modulating epigenetic processes. Despite the growing link between human brain disorders and the alteration of epigenetic state, there are few tools to directly probe epigenetic processes in vivo and none that can be used to probe epigenetic processes in the human brain. We feel new technologies for human molecular imaging that can report on enzymes which catalyze epigenetic transformations will revolutionize our ability to translate basic research to advances in human therapies. To address this critical need, our lab is developing radiotracers for positron emission tomography (PET) that can provide molecular-level epigenetic information about the human brain. Our group is working on a series of radiotracers for a number of epigenetic targets including histone deacetylases (HDACs) and lysine demethylases (KDMs). The presentation at BrainMap will detail our progress toward imaging HDACs and KDMs in the brain and highlight the potential of these agents to accelerate the discovery of therapeutics.

Alan Jasanoff, PhD
Associate Professor of Biological Engineering, MIT Depts. of Brain & Cognitive Sciences, Nuclear Science & Engineering, and McGovern Institute for Brain Research
AB '92 (biochemical sciences) Harvard College ; MPhil (chemistry) Cambridge University ; PhD '98 (biophysics) Harvard University. My laboratory is developing noninvasive functional imaging methods to study systems-level neural plasticity involved in low-level learning and perceptual behavior in small animals. We are seriously involved in the design of new imaging agents that may help define spatiotemporal patterns of neural activity with far better precision and resolution than current techniques allow. We have produced prototype imaging agents for "molecular fMRI," and are adapting them for application in vivo. Current imaging experiments in rodents focus on neural mechanisms involved in reward-related learning. In the past year we have developed an awake rat preparation which will allow us to image hemodynamic correlates (and eventually more direct measures) of brain activity while the animal performs tasks in an MRI scanner.

Functional molecular imaging in the brain
Functional magnetic resonance imaging (fMRI) with contrast agents sensitive to neural activity could have great impact in neuroscience by combining noninvasive whole-brain coverage with molecular-level specificity for neuronal events. Our research group is developing molecular fMRI approaches based on MRI-detectable sensors we have designed to monitor intra- and extracellular signalling events in the nervous system. Our sensors are built on a variety of chemical platforms, ranging from small molecules to nanoparticles. Protein-based contrast agents are of particular interest to us because of the possibility of gene-based brain delivery strategies and the availability of powerful protein engineering techniques. Here we describe the chemistry and bioengineering of several MRI sensors for neural activity, as well as the first efforts in our laboratory to perform functional neuroimaging with molecular specificity in living brains.

R. Todd Constable, PhD
Professor of Diagnostic Radiology, Neurosurgery, and Biomedical Engineering, Yale University Director, MRI Research; Co-Director of Yale MR Research Center
PhD, Medical Biophysics, University of Toronto, 1990. My research is focused on developing and validating novel approaches to functional Magnetic Resonance Imaging (fMRI) and using these methods to improve our understanding of brain function. This work includes approaches for quantitative neuroimaging and methods for assessing brain function via connectivity mapping. These developments are applied in the neurosurgical environment to localizing epileptogenic tissue and mapping function prior to surgical intervention. These studies provide a framework for validating the fMRI techniques through comparisons with cortical stimulation, behavioral analyses, Wada testing, and patient outcomes. They also improve our understanding of the link between fMRI signal changes and neuronal activity, through comparisons of fMRI in vivo with EEG?/?ERP recordings obtained in patients with depth electrodes and/or subdural grids. We are also interested in better understanding basic language and memory processing in humans and factors that influence the networks revealed by neuroimaging.

The Promise and Challenge of Voxel Based Measures of Intrinsic Connectivity
Resting-state functional magnetic resonance imaging has seen explosive growth in applications and interest in the past 5 years and while the promise of this approach is high, there remain challenges to it's widespread use in both neuroscience based research, clinical research and in clinical applications. This talk will review the pros and cons of task-based and task-free functional MRI studies with emphasis on the latter. Problems associated with seed based connectivity analysis and the application of network theory approaches to the analysis of resting-state fMRI data will be discussed. A novel voxel based approach for assessing intrinsic connectivity will be introduced that avoids the problems associated with seed based approaches and the arbitrary thresholds often used in such connectivity analyses.

Rick Hoge, PhD
Assistant professor in physiology and biomedical engineering at the Universite de Montreal, Canada Director, Laboratoire d'Imagerie NeuroVasculaire (LINeV)
Ph.D. at Mcgill University (1996-1999), post-doc at Harvard Medical School/Mass General Hospital (1999-2001), followed by an appointment as Instructor in Radiology at Harvard/MGH (2001-2006). Since 2006 the researcher has been an assistant professor in physiology and biomedical engineering at the Universite de Montreal and a laboratory director at the CRIUGM. Research interests: Quantitative physiological imaging using MRI and multimodal image fusion Role of physiological processes in supporting neuronal function Software architectures for analysis and visualization of functional neuroimaging data

Magnetic resonance imaging of vascular and metabolic function in human brain
Calibrated MRI methods based on hypercapnia or hyperoxia have been used previously to estimate fractional changes in oxidative metabolism elicited by a functional task. This talk will present recent developments by our group extending calibrated MRI methods to also measure resting cerebral oxygen metabolism in absolute micromolar units. The approach is based on a generalization of calibrated MRI models that unifies hypercapnic and hyperoxic approaches, and provides values for oxygen extraction fraction and resting oxygen consumption that are in excellent agreement with PET and whole-brain MRI values. General implications for BOLD fMRI will be discussed, as well as preliminary data on neurocognitive aging using hypercapnic MRI calibration.

Christophe Grova, PhD
Assistant Professor, Biomedical Engineering, McGill University, Canada Assistant Professor, Neurology & Neurosurgery, McGill University, Canada
From 1998 to 2002, Dr. Grova joined the IDM laboratory (UPRES 3192, University of Rennes 1, France), where he obtained a Ph.D. in "Validation of SPECT/MRI registration methods in the context of epilepsy". In January 2003 he joined the team of Jean Gotman at the Montreal Neurological Institute as a postdoctoral fellow. Dr. Grova investigates multimodal data fusion to characterize brain mechanisms and especially epileptic activity. His research project aims at developing methods to appropriately combine multimodal data in order to detect additional information that could be missed by considering each modality individually. A typical challenge is to combine modalities directly measuring neuronal activity with high temporal resolution with other modalities indirectly measuring the same function with high spatial resolution, through hemodynamic processes for instance.

Multimodal analysis of epileptic activity: insights from electrophyiology and hemodynamic measurements.
Interictal epileptic discharges, and notably, interictal spikes, are spontaneous neuronal discharges characteristic of the epilepsy of a patient. As opposed to seizures, these spontaneous events are not associated with clinical manifestations, thus allowing multimodal investigation. Such events could be detected using Electro- or Magneto-Encephalography (EEG /MEG) as large amplitude spontaneous events lasting around 100ms, that can be detected from physiological background activity. To be detectable from scalp data, it has been shown that the underlying generators of such epileptic activity should be spatially extended, a minimum area of 4 cm2 has been suggested in MEG and 6cm2 in EEG. In this context we proposed the Maximum Entropy on the Mean (MEM) framework to localize the generators of EEG/MEG activity together with their spatial extent. The first part of the talk introduced the MEM method and the evaluation of its performance when using EEG and then MEG data. A detailed comparison between inferences using entropic techniques and Restricted Maximum Likelihood in a hierarchical Bayesian framework will be presented. The second section of the talk introduced a time-frequency extension of the MEM framework, in order to localize oscillating activity in some specific frequency bands. This method we recently proposed has been validated using simulated data and applied to the localization of bursts of rhythmic epileptic activity. The third part of the talk will illustrate how these source localization techniques able to estimate the spatial extent of the generators could be used in a multimodal framework comparing electrophysiology and hemodynamic processes at the time of epileptic spikes. Illustration using EEG/fMRI daa and preliminary results using EEG/NIRS data will be presented.

Anne J Blood, PhD
Assistant Professor of Psychiatry at Harvard Medical School Research Scientist, Psychiatry at Massachusetts General Hospital Department of Neurology, MGH
PhD, Neuroscience, UCLA. My research involves using MRI to examine the pathophysiology of dystonia and other neurological movement disorders. Dystonia is a debilitating movement disorder, characterized by excessive, sustained involuntary muscle contractions, resulting in abnormal postures and impaired movement. The emphasis of my laboratory's research is on developing novel approaches to studying dystonia using functional and diffusion MRI techniques. For example, we are using "motor threshold" tasks to determine if the motor system is more easily engaged in dystonia patients than in people who do not have the disorder. We are also examining hemodynamic timecourses in motor and somatosensory brain regions during motor tasks to investigate how repetitive or skilled movements may contribute to development or exacerbation of brain pathology underlying dystonia. We have also recently begun using diffusion MRI to investigate whether there are mictrostructural abnormalities in the brains of dystonia patients. In the future, we plan to apply the approaches we've developed to study brain structure and function in other movement disorders, such as Parkinson's Disease.

Understanding the puzzle of motor control: A systems level approach

Yaoda Xu, PhD
Assistant Professor, Vision Sciences Laboratory, Psychology Department, Harvard University
Ph.D. in Cognitive Neuroscience, Dept. of Brain & Cognitive Sciences, MIT.


Seth Smith, PhD
Assistant Professor of Radiology and Radiological Sciences, Vanderbilt University Assistant Professor of Biomedical Engineering, Vanderbilt University Assisant Professor of Physics, Vanderbilt University Research Associate, F.M. Kirby Research Center, Kennedy Krieger Institute
Seth completed his Ph.D. in the Department of Biophysics and Biophysical Chemistry at Johns Hopkins. He is now assistant Professor at Vanderbilt University, and the Director of the Center for Human Imaging. His research is focused on developing novel MRI techniques for the study of the Brain and Spinal Cord in disease and in health. Notably Magnetization Transfer Imaging of the Brain and Spinal Cord and Iron Imaging.

Throwing the Box at it: Addressing the clinico-radiological paradox through advanced, quantitative MRI of small CNS structures
The clinical-radiological paradox is a recurring theme in the assessment of neurological damage in a variety of diseases. That is, the relationship between neurological dysfunction, nervous system repair and evolution are not always easily correlated with conventional brain MRI. Over recent years, non-conventional and quantitative MRI methods have been employed to address this paradox by increasing specificity for the pathological substrates of disease. We hypothesize that if we can characterize the damage to smaller structures of the central nervous system then it becomes more straightforward to relate these findings to neurological observations. The topic of this presentation will be to address how a quantitative MRI toolbox tailored to the study of the spinal cord and optic nerve can produce indices that can be related to specific neurological dysfunction. We will additionally examine the challenges of obtaining high resolution, quantitative data about the health and welfare of the tracts of the human spinal cord and optic nerves and how these impediments can be overcome.

Jorge Sepulcre, MD PhD
Harvard University
Jorge completed his neurology and PhD program in the Department of Neurology and of Neuroscience at the University of Navarra (2001 to 2004). In 2008, he joined the Athinoula A. Martinos Center for Biomedical Imaging, MGH, and the Harvard University Center for Brain Science as a Research Fellow in Prof. Randy Buckner’s lab. My work over the past four years, focused on the development of novel network analytical tools, has allowed me to take solid steps toward my overall research goal of applying such novel technologies to achieve better understanding of AD and other neurodegenerative diseases.

Large-Scale Functional Connectome of the Modal Brain
How human beings integrate information from external sources and internal cognition to produce a coherent experience is still not well understood. During the past decades, anatomical, neurophysiological and neuroimaging research in multimodal integration have stood out in the effort to understand the perceptual binding properties of the brain. Areas in the human lateral occipito-temporal, prefrontal and posterior parietal cortices have been associated with sensory bi- and trimodal processing (visual, auditory and somatosensory). Other subcortical brain regions, such as the superior colliculus, have been also involved in the integration of disparate modalities. Even tough this, rather patchy, organization of brain regions gives us a glimpse of the hierarchical sensory processing and the areas of perceptual convergence, the articulation of the flow of information from modality-related to the more parallel cognitive processing systems remains elusive. By using a novel technique called Stepwise Functional Connectivity (SFC) analysis, the present study analyzes the functional connectome and transitions from primary sensory cortices to higher-order brain systems. We identify a large-scale multimodal integration network and essential connectivity axes for perceptual integration in the human brain. The SFC analysis strategy reveals streamlining organizational principles of the modal and multimodal connectome.

Margaret Livingstone, PhD
Professor of Neurobiology at Harvard Medical School
Visiting Scientist in Radiology at Massachusetts General Hospital
PhD, Neurobiology, Harvard University

How do we acquire functional modules and what are they good for?
There are distinct regions of the brain, reproducible from one person to the next, specialized for processing the most universal forms of human expertise. What is the relationship between behavioral expertise and dedicated brain structures? Do reproducible brain structures mean only certain abilities are innate, or easily learned, or does intensive early experience influence the emergence of expertise and/or dedicated brain circuits? We found that intensive early, but not late, experience produces category-selective modules in macaque temporal lobe for stimuli never naturally encountered by monkeys, and produces more fluent processing of these stimuli than the same experience later in life. We propose that, as in early sensory areas, experience can drive functional segregation and that this segregation may determine how that information is processed.

Ciprian Catana, MD, PhD
Assistant Professor in Radiology at Harvard Medical School
Assistant in Neuroscience at Massachusetts General Hospital Department of Radiology, MGH
Ciprian Catana M.D., Ph.D. is the Director of Integrated MR-PET imaging at the Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, and Assistant Professor in Radiology at Harvard Medical School. Dr. Catana has extensive experience in combining PET and MRI. As part of his PhD work at University of California at Davis under the supervision of Dr. Simon Cherry, Dr. Catana designed and built an MR-compatible PET insert for a small animal 7-Tesla MRI system. He also performed the first in vivo animal experiments that demonstrated the ability to use this scanner for biomedical applications. After joining the Martinos Center, Dr. Catana has focused on translating this new imaging modality from the preclinical to the clinical arena and on developing quantitative MR-PET for human use.

When one plus one equals more than two - simultaneous MR-PET imaging of the brain
PET and MRI are two of the most powerful neuroimaging modalities. Recently, scanners capable of simultaneous PET and MR data acquisition in human subjects have become a reality and this new technology opens up possibilities impossible to realize using sequentially acquired data that could benefit many neurological applications. One such example is using the MR data for improving PET. While PET as a technique has many advantages, its accuracy is confounded by several factors. For example, attenuation and scatter correction have to be performed to account for the interactions of the gamma-ray photons in the subject before reaching the detectors; motion correction has to be applied to avoid the degradation of the images due to involuntary head movements; partial volume effect correction is required due to the relatively limited spatial resolution; an input function is required for accurate estimation of parameters of interest. The spatially and temporally correlated MR data acquired simultaneously offer the unique opportunity to correct for these confounding effects and improve the reliability and reproducibility of the PET estimates. On the other hand, the temporal correspondence of PET signals might help us better understand a number of MR techniques in vivo. In this talk, we will discuss our progress on implementing and validating these methods and our initial experience using this novel imaging technique for neuroimaging studies.

Eng H. Lo, PhD
Professor in Radiology at Harvard Medical School Associate Biochemist in Neurology at Massachusetts General Hospital
Department of Radiology, MGH
Dr. Lo received his B.S. in electrical engineering from Yale, his Ph.D. in biophysics from the University of California Berkeley, and completed his postdoctoral training in neuroscience at Stanford. He moved to MGH in 1991, and is currently the Head of the Neuroprotection Research Lab, and Professor of Neuroscience in Harvard Medical School.

Neurovascular mechanisms of injury and repair after stroke
Over the past decade, numerous advances in neuroimaging have allowed us to probe the pathophysiology of brain injury after stroke. MRI tools centered on diffusion-perfusion mismatch may help identify patients with salvageable penumbra. PET imaging may help us characterize the ensuing dysregulations in cerebral blood flow and metabolism. However, in spite of these powerful in vivo techniques for "looking at stroke", we still lack clinically effective neuroprotective therapies. In this presentation, we will attempt to discuss the translational challenges involved in bridging promising experimental leads into clinically meaningful applications. Specifically, we try to address the following 3 concepts: Is it possible that beyond saving neurons (i.e. neuroprotection per se), we need to consider restoring cell-cell interactions between multiple brain cell types (i.e. the neurovascular unit)? Is it possible that many of the neurovascular mechanisms and targets underlying stroke are biphasic in nature (i.e. deleterious in the acute stage but beneficial during recovery)? Finally, in addition to cell and animal models, is it possible to develop "human models" that may help us link experimental platforms to the stroke patient?

Garth M. Beache, MD
Associate Professor, Department of Radiology, School of Medicine, University of Louisville
M.D., Howard University College of Medicine, Washington, DC The Science of Clinical Investigation (1-yr Certificate Program) The Johns Hopkins University School of Medicine & Public Health and Office of Continuing Medical Education

Imaging Signaling Pathways in the Insulin Cardiometabolism Syndrome: Mechanics-Vascular Coupling
The overall goal of this research is to define magnetic resonance markers that characterize a putative underlying abnormality of small vessel regulation, and energetically-dependent final-common-pathway functioning in a related group of conditions that are linked to heart microvascular disease. This has potential implications for therapeutic interventions targeted to biological mechanisms in these diseases. This work has parallels for researches in the brain.

Ed Boyden, PhD
Benesse Career Development Professor
Leader, Synthetic Neurobiology Group
Associate Professor, MIT Media Lab
Joint Professor, MIT Dept. of Biological Engineering, MIT Dept. of Brain and Cognitive Sciences
Investigator, MIT McGovern Institute
Associate Member, MIT Picower Institute
Ed Boyden is the Benesse Career Development Professor, and Associate Professor of Biological Engineering and Brain and Cognitive Sciences, at the MIT Media Lab and the MIT McGovern Institute. He leads the Synthetic Neurobiology Group, which develops tools for controlling and observing the dynamic circuits of the brain, and uses these neurotechnologies to understand how cognition and emotion arise from brain network operation, as well as to enable systematic repair of intractable brain disorders such as epilepsy, Parkinson's disease, post- traumatic stress disorder, and chronic pain. The tools his group has invented include a suite of 'optogenetic' tools that are now in use by hundreds of groups around the world, for activating and silencing neurons with light. These tools enable the causal assessment of how specific neurons contribute to normal and pathological brain functions, revealing with great temporal precision the processes for which their activities are necessary or sufficient. He has launched an award-winning series of classes at MIT that teach principles of neuroengineering, starting with basic principles of how to control and observe neural functions, and culminating with strategies for launching companies in the nascent neurotechnology space.

Optogenetics, Robotic Neural Recording, and Other Neuroscience Tools
Understanding how neural circuits implement brain functions, and how these computations go awry in brain disorders, is a top priority for neuroscience. Over the last several years we have developed a rapidly-expanding suite of genetically-encoded reagents that, when expressed in specific neuron types in the nervous system, enable their electrical activities to be powerfully and precisely activated and silenced in response to pulses of light. These tools are in widespread use for analyzing the causal role of defined cell types in normal and pathological brain functions. In this talk I will briefly give an overview of the field, and then I will discuss a number of new tools for neural activation and silencing that we are developing, including new molecules with augmented amplitudes, improved safety profiles, novel color and light-sensitivity capabilities, and unique new capabilities. We have begun to develop hardware to enable complex and distributed neural circuits to be precisely controlled, and for the network-wide impact of a neural control event to be measured using distributed electrodes, fMRI, and robotic intracellular neural recording. We explore how these tools can be used to enable systematic analysis of neural circuit functions in the fields of emotion, sensation, and movement, and in neurological and psychiatric disorders. Finally, we discuss our pre-clinical work on translation of such tools to support novel ultraprecise neuromodulation therapies for human patients.

Keith A. Johnson, MD
Associate Professor of Radiology and Neurology, Harvard Medical School
Associate Radiologist and Director of Molecular Neuroimaging in the Division of Nuclear Medicine and Molecular Imaging, MGH
MD, University of Kansas School of Medicine
Residency, Brigham and Women's Hospital
Fellowship, Massachusetts General Hospital

Molecular, structural and functional imaging in preclinical Alzheimer's disease
Converging evidence suggests that the pathophysiological process of Alzheimer’s disease (AD) begins more than a decade before the clinical stage we now recognize as AD dementia. Even by the stage of mild impairment, the neurodegeneration of AD is thought to be well entrenched, and amyloid deposition has already been present for years. Attempts to clarify the pathogenetic chain of events have proceeded in parallel with the related task of human AD biomarker development. In this talk, I will review recent data from normal older adults in which amyloid deposition is related to brain volume and brain glucose metabolism, in an attempt to characterize the preclinical state and to develop therapeutic trial endpoints.


Marjanska Malgorzata, PhD
CMRR, University of Minnesota
Research Assistant Professor Malgorzata (Gosia) Marjanska received a BS in Chemistry from Loyola University of Chicago and PhD also in Chemistry from University of California at Berkeley. During her PhD, she worked on different NMR methods with the use of liquid crystals as an ordering medium. She came to work at CMRR at the University of Minnesota as a post-doc in 2002 and stayed on as Research Assistant Professor. She has published 28 journal articles lately focusing on methods and applications of MR spectroscopy.

Hyperpolarized 13C Spectroscopy in Rat Brain at 9.4 T and Localized 1H Spectroscopy in Human Brain at 7 T
Carbon-13 spectroscopy combined with the infusion of 13C-labeled substrates is a powerful tool to study brain metabolism in vivo. Detection of hyperpolarized [1-13C]pyruvate and its metabolic products has been reported in vivo in rats, mice, pigs, and in isolated rat hearts, and [2-13C]pyruvate in isolated rat hearts. In my presentation, I will focus on our experience with measuring 13C signals of hyperpolarized 13C metabolic products in the rat brain in vivo following the injection of hyperpolarized [1-13C]pyruvate and [2-13C]pyruvate. A metabolic model will be presented to fit hyperpolarized lactate and bicarbonate 13C time courses measured after [1-13C]pyruvate injection. Proton magnetic resonance spectroscopy offers a non-invasive way to quantify metabolites in vivo. Metabolite levels are sensitive to different in vivo pathologic processes at the molecular or cellular level. At ultra high field strengths, neurochemical profile with up to 18 metabolites can be quantified in vivo. In my presentation, I will discuss the efficacy of using the LASER sequence to obtain high-quality 7-T spectra from different brain regions, to measure T2 relaxation times, and to quantify concentrations of metabolites in different brain regions.

David C. Alsop, PhD
Professor, Department of Radiology, Harvard Medical School
Staff Ph.D., Radiology, Beth Israel Deaconess Medical Center
David Alsop, PhD, is Director of MRI Research in the Department of Radiology at Beth Israel Deaconess Medical Center and Professor of Radiology at Harvard Medical School. His research has focused on technological advancements in MRI and their translation to research and clinical applications. A particular focus is on developing techniques to provide new forms of contrast in MRI in order to better characterize disease. He is perhaps best know for his work on arterial spin labeling perfusion MRI, especially techniques for labeling and acquisition.

Advanced Techniques and Applications of Perfusion MRI
Arterial spin labeling perfusion MRI provides a highly stable and quantitative functional measure that can open new designs for functional studies of the brain. Cross-sectional studies of disease or genetic variants can provide important insights. The noninvasiveness of ASL make it particularly suited to studies of disease progression and treatment, and to pharmacologic or other physiologic perturbations. Some recently developed methods for improving sensitivity, quantification, and speed of acquisition will be presented. Applications to pharmacologic studies, aging and dementia, progression and treatment of cancer, and to the study of resting-state fluctuations of the brain will be described.

Gitte M Knudsen, MD, PhD
Chairman of Neurobiology Research Unit and of Center for Integrated Molecular Brain Imaging, Rigshospitalet and University of Copenhagen, Denmark
Visiting Scientist, MGH Martinos Center
PhD, Neuroscience, University of Copenhagen
MD, Medical School, University of Copenhagen

Molecular Brain Imaging of the Serotonergic Transmitter System
The serotonergic system plays a key modulatory role in the brain and is the target for many drug treatments for brain disorders either through reuptake blockade or via interactions at the 14 subtypes of serotonin (5-HT) receptors. The talk will give a status of radioligands used for positron emission tomography (PET) imaging of human brain serotonin (5-HT) receptors and the 5-HT transporter (SERT). Currently available radioligands for in vivo brain imaging of the 5-HT system in humans include PET radiotracers for the 5-HT1A, 5-HT1B, 5-HT2A, and 5-HT4 receptors, and for SERT. Attempts to develop a radiotracer probe for in vivo imaging of the spatial and temporal release of 5-HT are underway. Examples of clinical applications of in vivo brain imaging of the 5-HT system, ie, in mood disorders and Alzheimers Disease will also be given.

Kawin Setsompop, PhD
MGH Martinos Center
Kawin graduated from the University of Oxford, U.K. where he obtained his Master degree in Engineering and science. He then acquired his PhD in Electrical Engineering and Computer Science at the Massachusetts Institute of Technology in 2008, under the supervision of Professor Elfar Adalsteinsson. At present, Kawin is an instructor in Radiology at the Martinos Center for Biomedical Imaging working in Professor Lawrence Wald's research group.

Efficient Diffusion Imaging Acquisition
In this work, we demonstrate an improvement in the time efficiency of diffusion acquisitions using three complementary technologies: (i) high-strength gradient coils, (ii) a Simultaneous Multi-Slice (SMS) acquisition with Blipped-CAIPI readout using a highly parallel receive coil array, and (iii) a compressed sensing reconstruction that enables undersampling of q-space. Together, these improvements allow for an acquisition of high-quality whole-brain DSI data in just 4 minutes. While this initial demonstration focuses on DSI, the general approach should be applicable to other HARDI acquisition schemes.

Michael Esterman, PhD
Boston University School of Medicine
Associate Director of the VA Boston Neuroimaging Center
Michael Esterman is a co-founder of the Boston Attention and Learning Lab. He received his degree in cognitive psychology at UC Berkeley, where he investigated spatial attention and object perception using transcranial magnetic stimulation (TMS). In his post-doctoral fellowship at Johns Hopkins University, he investigated the neural mechanisms of cognitive control, with an emphasis on using pattern classification to decode attentional states. He is now an Assistant Professor at the Boston University School of Medicine, and Associate Director of the VA Boston Neuroimaging Center. Mike's current interests include investigating the neural basis of attentional control and distractibility, in both healthy young and old adults, as well as in patients with PTSD, TBI, and focal brain injury.

In the zone or zoning out? Tracking behavioral and neural fluctuations during sustained attention
Despite growing recognition that intrinsic brain activity persists during cognitive performance and influences behavior, prevailing fMRI analysis strategies involve averaging data across multiple trials or time points, treating moment-to-moment fluctuations as noise. Using alternative approaches, we clarify the relationship between ongoing brain activity and performance fluctuations during sustained attention. We introduce a novel task [the gradual onset continuous performance task (GO-CPT)], along with innovative analysis procedures that probe the relationships between reaction time (RT) variability, attention lapses, and intrinsic brain activity. Our results highlight two distinct attentional states - a stable, less error-prone state characterized by higher default mode network (DMN) activity but during which subjects are at risk of erring if DMN activity rises beyond intermediate levels, and a more effortful mode of processing that is less optimal for sustained performance and relies on activity in dorsal attention network (DAN) regions. These findings motivate a new view of DMN and DAN functioning capable of integrating seemingly disparate reports of their role in goal-directed behavior. Further, they hold potential to reconcile conflicting theories of sustained attention, and represent an important step forward in linking intrinsic brain activity to behavioral phenomena.

Dylan Tisdall, PhD
MGH Martinos Center
Dylan Tisdall is a Research Fellow at the Athinoula A. Martinos Center for Biomedical imaging. He received an Hons. BMath from University of Waterloo (Waterloo, Ontario, Canada), and his Ph.D. in Computing Science from Simon Fraser University (Burnaby, British Columbia, Canada), before joining the Martinos Center in 2008. His main research interests are pulse sequence design, image reconstruction, and statistical signal processing methods for quantitative MRI analysis. His major projects at Martinos have been in motion-corrected sequence for morphometry and sequence development for the Connectome scanner.

Quantifying the Connectome Scanner
The MGH/UCLA Connectome scanner, now in Bay 8 at Charlestown Navy Yard, is a 3T whole-body human MRI scanner with one-of-a-kind gradients designed to provide better SNR for diffusion imaging than any human system ever produced. The principal difference between the Connectome scanner and a "normal" 3T system lies in the maximum amplitude that the gradient system can achieve --- 7.5 times greater than the gradients on the Trio and Avanto systems. This allows a dramatic shortening of the echo time, and in consequence a dramatic increase in the image SNR when performing diffusion-weighted MRI. These novel high-amplitude gradients also come with downsides (concomitant fields, eddy currents) that that must be accounted for to make the scanner usable. In this talk, I will discuss the ongoing work at MGH to quantify both the improvements in SNR and the behaviour of the eddy currents using novel analysis techniques and equipment.

Vitaly Napadow, PhD
MGH Martinos Center
Vitaly Napadow is an assistant professor at the Martinos Center for Biomedical Imaging at Massachusetts General Hospital and Harvard Medical School in Boston, MA. Vitaly holds a secondary appointment as assistant professor in the Pain Management Center at Brigham and Women's Hospital. He received his Ph.D. in biomedical engineering from the Harvard-MIT Health Sciences and Technology program and a Masters degree from the New England School of Acupuncture (NESA). His research expertise is in MRI neuroimaging and his interests focus on evaluating brain processing underlying aversive perceptual states such as chronic pain, itch, and nausea, as well as central mechanisms supporting potential therapies such as acupuncture.

Neuroimaging markers for chronic pain disorders - objective outcomes for evaluating acupuncture therapy
Neuroimaging has show promise in providing objective biological markers that track subjective clinical symptomatology such as chronic pain. The development of such markers will aid in identifying objective outcomes for use in clinical trials and can help inform both the underlying mechanisms for different chronic pain conditions and potential mechanisms of action supporting different therapies. Our lab has been developing potential fMRI markers for two different chronic pain conditions: carpal tunnel syndrome, which is defined by a focal peripheral nerve lesion, and fibromyalgia, which is a global pain disorder. These markers include altered S1 somatotopy for CTS, which tracks with peripheral nerve conduction latencies; and altered resting brain connectivity for fibromyalgia, which tracks with clinical pain intensity. Moreover, our proposed markers are sensitive to treatment by acupuncture, an ancient Eastern therapy that has been receiving growing acceptance by the conventional Western healthcare system. Acupuncture has been found to reduce symptomatology in both CTS and fibromyalgia, and may normalize the neuroimaging markers defined above. Our findings both inform our understanding of the central neural mechanisms underlying chronic pain disorders, and provide more objective outcome metrics for evaluating acupuncture therapy.

Jason Bohland, PhD
Boston University
Jason Bohland is a computational neuroscientist, whose research program involves a combination of computational, informatics, and experimental methods to study problems of brain architecture, often focusing on speech and language brain systems. He received degrees in computer and electrical engineering from the University of Cincinnati and his PhD in Cognitive and Neural Systems from Boston University in 2007, working with Dr. Frank Guenther in brain imaging and computational modeling of speech sequence planning and production. He served as a postdoctoral fellow and research scientist at Cold Spring Harbor Laboratory, working with Dr. Partha Mitra within the Brain Architecture Project before returning to BU as faculty in 2009.

Data-driven studies of large-scale molecular and circuit architecture of the brain
As large brainwide datasets become increasingly available, it is now possible to ask fundamental questions about brain organization that were previously unapproachable. In particular, spatial gene expression profiles, such as those made available in the Allen Brain Atlases, hold the promise to allow fusion of information from the molecular level with functional and anatomical data represented as brain maps. In this talk I will focus primarily on exploratory, multivariate analyses of these gene expression atlases in mouse and human. I will describe correlations across the spatial expression profiles of a large (N>3000) set of genes, as well as spatial autocorrelation in gene expression leading to clusters of voxels that largely mirror classically-defined neuroanatomical regions. I will further discuss integrating these data in the study of heritable disorders, including recent preliminary work in speech and language disorders. I will also very briefly describe the objectives and progress on the collaborative Mouse Brain Architecture Project, which aims to generate brainwide maps of inter-regional ("mesoscopic") neural connectivity in the mouse. This portion of the talk will focus primarily on the informatics challenges involved in a project of this scale.

Daniel Gochberg, PhD
Vanderbilt University
Department of Biomedical Engineering, Vanderbilt University, Nashville, Tennessee, USA
Institute of Imaging Science, Vanderbilt University, Nashville, Tennessee, USA

CERT vs. CEST: a New Approach to Imaging Amide Exchange
Chemical exchange saturation transfer (CEST) has recently received significant attention as a way of imaging solutes, and especially proteins, that contain amides. This imaging method has applications in imaging protein content in cancer studies, and in imaging amide-water exchange rates, and hence pH, in stroke studies. However, these studies have been unoptimized, rarely quantitative, and susceptible to artifacts originating in signal contributions from non-amide sources, such as asymmetric background macromolecules, static field inhomogeneities, and overlapping amine resonances. We will introduce a new approach based on separating magnetization saturation and rotation that addresses many of the shortcomings of CEST amide imaging.

Darin Dougherty, MD
MGH Martinos Center
Dr. Dougherty received his MD from the University of Illinois and completed his residency in general psychiatry at Massachusetts General Hospital. He is a graduate of the Clinical Investigator Training Program at Harvard Medical School and Massachusetts Institute of Technology. Dr. Dougherty's research interests include neuroimaging, neurobiology, psychopharmacology, and neurotherapeutic (device and/or surgical) interventions for the treatment of treatment-refractory psychiatric illness.

Neurotherapeutic Interventions in Psychiatry
While the majority of patients with psychiatric illness respond to conventional treatments (e.g., psychotherapy, pharmacotherapy), a minority do not and are considered treatment-resistant. For these patients, neurotherapeutic interventions may be warranted. Neurotherapeutic interventions refer to surgical and/or device-related treatments and include electroconvulsive therapy (ECT), transcranial magnetic stimulation (TMS), ablative limbic system surgery (e.g., anterior cingulotomy, limbic leukotomy), epidural cortical stimulation (EpCS), and deep brain stimulation (DBS). I will review our experience with each of these procedures at MGH and will include results from concurrent imaging studies regarding the pathophysiology of these illnesses, changes associated with these interventions, and the potential utility of using pretreatment neuroimaging to predict subsequent response. I will conclude with a review of possible future directions within the field of psychiatric neurotherapeutics including targeted delivery of neurotransmitters or neurotrophic growth factors and optogenetics.