Normal Brain Physiology by BOLD

Thursday 15 May 2014

Alex Bhogal¹, Marielle Philippens², Jeroen C.W. Siero¹, Esben Petersen¹, Joe Fisher³, Peter Luijten¹, and Hans Hoogduin^1
1Radiology, University Medical Center, Utrecht, Netherlands, ²Radiotherapy, University Medical Center, Utrecht, Netherlands, ³Anaesthesia & Physiology, University Health Network, Toronto, Ontario, Canada

Here we examine differences in the cerebrovascular reactivity (CVR) response of gray and white matter to a progressive hypercapnic stimulus. Tissue-specific differences arise which cannot be explained by CVR response delay alone.

Zachary B Rodgers¹, Erin K Englund¹, Michael C Langham¹, and Felix W Wehrli^1
1Radiology, University of Pennsylvania, Philadelphia, PA, United States

Susceptometry-based oximetry (SBO) and T₂-Relaxation-Under-Spin-Tagging (TRUST) are two promising methods for global CMRO₂ quantification. However, while SBO offers high temporal resolution, it is sensitive to vessel geometry and requires robust background field removal. TRUST does not have such limitations, but suffers from inherently low temporal resolution and must quantify flow from a separate PC-MRI measurement. By inserting a flow-encoded, SBO pulse sequence within the T₁ recovery period of TRUST and using fewer effective echo times for T₂ fitting, we demonstrate that it is possible to achieve T₂-based CMRO₂ quantification in only seconds, while directly comparing SBO- and T₂-derived Y_v.

Audrey P Fan^1,2, Karleyton C Evans^1,3, Jeffrey N Stout^2,4, Bruce R Rosen^2,4, and Elfar Adalsteinsson^1,4
1Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA, United States, ²Radiology, Athinoula A. Martinos Center for Biomedical Imaging, Charlestown, MA, United States, ³Psychiatry, Massachusetts General Hospital, Boston, MA, United States, ⁴Health Sciences and Technology, Harvard-MIT, Cambridge, MA, United States

The purpose of the study was to validate regional measurements of oxygen extraction fraction (OEF) in individual cerebral veins derived from quantitative susceptibility mapping. Flow-compensated gradient echo images and arterial spin labeling perfusion images were acquired during eucapnia and moderate hypercapnia in eight healthy volunteers. Regional changes in perfusion were used to predict local oxygenation changes and compared with quantitative susceptibility mapping (QSM)-based OEF estimates in veins draining the cingulate, occipital, parietal, and frontal cortices. Measured OEF changes during hypercapnia from MR susceptibility show good agreement with predictions based on regional flow observations and no change in the cerebral metabolic rate of oxygen CMRO₂, and suggests that regional estimates of OEF in vessels from QSM maps are reliable.

Peiying Liu¹, Yang Li¹, and Hanzhang Lu^1
1Advanced Imaging Research Center, University of Texas Southwestern Medical Center, Dallas, Texas, United States

Cerebrovascular reactivity (CVR), measured by combining BOLD MRI and hypercapnia, is an important marker of cerebrovascular function. An undesirable feature of current CVR mapping results is that the image shows very strong contrast in large drains due to large blood volume in these voxels, which reduces the conspicuity of true CVR abnormalities. The present work demonstrates that normalizing CO2-CVR by O2-reactivity can effectively reduce the influence of cerebral blood volume on the results. Furthermore, a novel breathing paradigm with concomitant CO2/O2 challenge is developed to obtain CO2 and O2 reactivity maps in a single scan without increasing the study duration.

Felix Schlegel¹, Aileen Schroeter¹, and Markus Rudin^1,2
1Institute for Biomedical Engineering, ETH and University of Zürich, Zürich, Zürich, Switzerland, ²Institute of Pharmacology and Toxicology, University of Zürich, Zürich, Zürich, Switzerland

For stimulus-evoked fMRI in mice, the standard analysis tools may not be appropriate as the intrinsic assumptions of linearity relating stimulus and hemodynamic response and the use fixed shapes of the hemodynamic response function (HRF) may not be valid. Here, we show how the sensitivity and specificity of BOLD fMRI in mice can be improved by employing appropriate experimental designs and analysis approaches. We found that in out data, event-related designs produce more specific responses than block designs and that nonlinear regression approaches outperform traditional methods like finite impulse responses, SPM basis functions or fixed HRF shapes.

ChiaChun Hung^1,2, Cecil Chern-Chyi Yen¹, Jennifer L Ciuchta¹, Daniel Papoti¹, David A Leopold², and Afonso C Silva^1
1CMU/LFMI/NINDS, National Institutes of Health, Bethesda, MD, United States, ²SCNI/LN/NIMH, National Institutes of Health, Bethesda, MD, United States

The ventral temporal cortex of humans and macaques is marked by multiple, distinct regions specialized for the visual processing of faces and bodies. Here we demonstrate a similar category-specific organization in the common marmoset (Callithrix jacchus), a small, New World monkey. By using functional magnetic resonance imaging (fMRI) and electrocorticography (ECoG) in awake, trained animals, we demonstrate six face-selective and three adjacent body-selective cortical regions in the marmoset visual cortex. These findings suggest that key elements of ventral stream specialization for face and body processing were present in a common anthropoid primate ancestor at least 35 million years ago.

Anna Gaglianese^1,2, Peter Herman¹, Daniel Coman¹, Pietro Pietrini^2,3, and Fahmeed Hyder^1,4
1Department of Diagnostic Radiology - Magnetic Resonance Research Center, School of Medicine, Yale University, New Haven, CT, United States,²Laboratory of Clinical Biochemistry and Molecular Biology, Department of Surgical, Medical, Molecular Pathology and Critical Area, University of Pisa, Pisa, Italy, ³Clinical Psychology Branch, Pisa University Hospital, Pisa, Italy, ⁴3Department of Biomedical Engineering, School of Engineering and Applied Science, Yale University, New Haven, CT, United States

Characterizing the relationship between neurovascular and neurometabolic changes that occurs in the brain is crucial for fMRI measurements. Here, we developed a model that relates changes in temperature in space and time to layer specific variations of CBF, CBV and BOLD acquired with differente tecniques and CMR₀₂ measured by calibrated MRI, under rest and stimulation. We estimated and measured temperature changes nearly 0.1 °C in the middle/lower layers in agreement with a tighter coupling between CBF and CMR₀₂ during stimulation. Temperature mapping and fMRI measurements may provide an useful tool to measure metabolic and hemodynamic variations in the brain.

Sally Eldeghaidy^1,2, Luca Marciani³, Joanne Hort⁴, Tracey Hollowood⁴, Gulzar Singh⁵, Debbie Bush⁶, Tim Foster⁷, Andy J. Taylor⁴, Johanneke Busch⁸, Robin C. Spiller³, Penny A. Gowland², and Susan T. Francis^2
1Department of Physics, Faculty of Science, Suez Canal University, Ismailia, Egypt, ²Sir Peter Mansfield Magnetic Resonance Centre, University of Nottingham, Nottingham, United Kingdom, ³Nottingham Digestive Diseases Centre, University of Nottingham, Nottingham, United Kingdom, ⁴Flavour Research Group, Division of Food Sciences, University of Nottingham, Nottingham, United Kingdom, ⁵School of Biomedical Sciences, University of Nottingham, Nottingham, United Kingdom, ⁶Division of Surgery, Queen's Medical Centre University Hospital, Nottingham, United Kingdom, ⁷Division of Food Sciences, University of Nottingham, Nottingham, United Kingdom, ⁸Unilever Food and Health Research Institute, Unilever R&D, Vlaardingen, Netherlands

We investigated the modulation of BOLD responses to fat and control samples following prior feeding of a fat and water meal. The BOLD response in the anterior insula following consumption of the fat meal was suppressed compared to the water meal, with an increased rate of habituation to samples. Significant habituation was also seen in the amygdala, mid- and posterior insula. A decrease in reward and oral somatosensory activity was seen with increasing CCK and fullness values. These changes were found to be associated with a baseline reduction in CBF in the hypothalamus, thalamus and insula following the fat meal.

Fani Deligianni¹, Maria Centeno¹, David W. Carmichael¹, and Jonathan D. Clayden^1
1Institute of Child Health, UCL, London, United Kingdom

It has been shown that the resting-state (rs) networks observed with fMRI reflect electrophysiological activity. However, questions remain regarding which EEG features most closely reflect rs-fMRI networks. We relate the covariance structure of the envelope of the source localised electrophysiological signal to the covariance derived from rs-fMRI. We use statistical inference to predict EEG brain connectivity from fMRI connectivity and vice-versa. Our results support that the β band EEG activity best reflects spontaneous cognitive operations during conscious rest. Furthermore, fMRI connectivity can be predicted from EEG in any band indicating that there are signatures of rs-fMRI dynamics across EEG frequencies.

Yuzheng Hu¹, Xi Chen¹, Hong Gu¹, and Yihong Yang^1
1Neuroimaging Research Branch, National Institute on Drug Abuse, NIH, Baltimore, MD, United States

The default mode network (DMN) deactivation observed with BOLD fMRI is thought to be associated with suppression of spontaneous brain activities and reallocation of resources to ongoing, attention-demanding tasks. However, the underlying neurochemical mechanism of DMN deactivation remains largely unknown. The current study aimed to examine the relationship between DMN deactivation induced by a WM task and the endogenous concentrations of glutamate and GABA in the posterior cingulate/precuneus region. The major excitatory neurotransmitter, glutamate, was found to prevent BOLD signal from deactivation while GABA, the major inhibitory neurotransmitter, exert opposite effects.