Pediatr Blood Cancer 2014;61:1724–1728
REVIEW Magnetic Resonance Imaging in the Evaluation of Cognitive Function Erin D. Bigler, Image quality of magnetic resonance imaging (MRI) scans of the brain currently approximate gross anatomy as would be viewed at autopsy. During the first decade of the 21st Century incredible advances in image processing and quantification have occurred permitting more refined methods for studying brain-behaviorcognitive functioning. The current presentation overviews the current status of MRI methods for routine clinical assessment of brain
PhD
1,2,3,4
*
pathology, how these techniques identify neuropathology and how pathological findings are quantified. Diffusion tensor imaging (DTI), functional MRI (fMRI), and resting state fMRI are all reviewed, emphasizing how these techniques permit an examination of brain function and connectivity. General regional relationships of brain function associated with cognitive control will be highlighted. Pediatr Blood Cancer 2014;61:1724–1728. # 2014 Wiley Periodicals, Inc.
Key words: cognitive function; diffusion tensor imaging; functional magnetic resonance imaging; magnetic resonance imaging
INTRODUCTION The role of magnetic resonance imaging in assessing cognitive function in the neurological and/or neuropsychiatric patient has a relatively short history because the ability for in vivo noninvasive imaging of the brain has only been available since the 1970s [1]. Although by the 1980s computed tomography (CT) had become established along with the beginnings of what was then called nuclear magnetic resonance imaging, the forerunner to contemporary magnetic resonance (MR) imaging or MRI, the image quality of the brain was limited as shown in Figure 1. Advancement in image clarity and presentation dominated neuroimaging through the end of the 20th century [2] with ever improving image display as also shown in Figure 1. While in the early days of brain imaging studying brain-cognitive-behavior correlates were problematic because of the coarseness of the images and the limits of how pathology could be identified [3], by the beginning of the 21st century thin-slice MRI approximated what could be viewed at post-mortem as shown in Figure 1. With issues of image quality now achieved, the next advancements occurred in image quantification. Once more refined image quantification was achieved, an explosion in the study of MRI findings and cognition occurred [4].
TISSUE SEGMENTATION At a gross morphological level, three main characteristics of the brain can be identified—white matter, gray matter, and cerebrospinal fluid (CSF) filled spaces and cavities. There are also neuroimaging methods that permit viewing the cerebrovasculature, but in normal brain parenchyma the small vessels and all capillaries embedded within tissue often cannot be differentiated from the white or gray matter wherein they are embedded. The above biological facts provide the basis for a simple classification in neuroimaging where white matter, gray matter, and CSF are delineated in what is referred to as a segmented image as shown in Figure 2C. By identifying a specific color for each tissue type or region of interest (ROI), a color-coded classified image may be generated which in turn provides a method for image quantification since MRI data (signal intensity) used to generate a brain image has a slice thickness and a known distance from the next
C
2014 Wiley Periodicals, Inc. DOI 10.1002/pbc.25110 Published online 11 June 2014 in Wiley Online Library (wileyonlinelibrary.com).
slice. Knowing these metrics provides the basis for volumetric measurements. For example, in Figure 2 from a child who sustained a severe traumatic brain injury (TBI), the different MR sequences were used to isolate the total amount of destructive damage to the cerebral cortex, as shown in red. These areas of damage relate to cognitive deficits in executive and emotional functioning.
TISSUE CLASSIFICATION The next step in image quantification came with classifying anatomical regions. A variety of automated image analysis methods capitalized on key landmarks which provided a roadmap for anatomical identification. Using one type of automated image classification method referred to as FreeSurfer [5] permits the identification and quantification of both cortical and subcortical ROIs. The effectiveness of this technique is that it is reliable and automated making the process ideal for analyzing large numbers of scans within a clinical population compared to a control sample [6,7]. With these types of quantitative measures, relations between certain brain regions and cognition could now be examined [8]. Likewise, pathology can be shown in threedimensional (3D) space as depicted in Figure 3. This child also had a severe TBI with major volume loss involving the left frontal lobe. As in Figure 2, the volume loss is depicted in red, but this time the deficit is in the left frontal area. This type of image presentation can be used to show any type of pathology from degenerative to vascular or neoplastic. But even more importantly, automated programs like FreeSurfer permits the
1
Department of Psychology, Brigham Young University, Provo, Utah; Neuroscience Center, Brigham Young University, Provo, Utah; 3 Department of Psychiatry, University of Utah, Salt Lake City, Utah; 4 The Brain Institute of Utah, University of Utah, Salt Lake City, Utah 2
Correspondence to: Erin D. Bigler, Department of Psychology and Neuroscience Center, 1001 SWKT, Brigham Young University, Provo, UT 84602. E-mail: [email protected] Received 22 April 2014; Accepted 23 April 2014
MRI and Cognitive Function
1725
Fig. 1. Brief history of brain neuroimaging: (A) plain skull film in use since the early 1900s; (B) pneumoencephalography that provided a silhouette of the ventricle after cerebral spinal fluid (CSF) had been replaced with air in use through most of the 1970s; (C) early computed tomography (CT) scan circa 1975; (D) contemporary CT; (E) prototype magnetic resonance imaging (MRI) from the late 1970s; and (F) contemporary 3 Tesla MRI.
volume quantification of all cortical gyri (both white and gray matter along with cortical thickness) as well as all sub-cortical structures. For example, in Figure 3, not only can the cortical gyri be viewed but also the volumes calculated for the hippocampus, amygdala, basal ganglia, ventricular system, midbrain, and brainstem are shown. Once isolated like this, the
potential role of these ROIs in the target outcome behavior can be analyzed. Another approach that capitalizes on the segmented image is to “normalize” each brain within the same 3D space. By using small voxels based on the segmented image that are either white, gray or CSF, a concentration count for each of these tissue types can
Fig. 2. Focal right frontal lesion in traumatic brain injury (TBI) shows differences depending on the magnetic resonance (MR) image sequence. The focal cortical lesions are obvious in the T1 and T2 images (see arrows) but in the fluid-attenuated inversion-recovery (FLAIR) sequence visible extensive white matter differences are evident when comparing the right frontal area (arrow) with the left. The arrow in the gradient-recalled echo (GRE) sequence points to dark, hypointense signal differences that reflect old blood. In the diffusion tensor imaging (DTI) image at this same level note the loss of the distinctiveness of the green colorization which signifies a loss of white matter integrity throughout the right frontal lobe. Green represents pathways that course front to back (anterior to posterior), blue represents vertically oriented tracts and warm colors (orange-red) represents side-to-side coursing tracts. All of these abnormalities can be projected onto a 3D image of the brain as shown in (A). Typically radiological convention is to show the MRI as if looking at the patient so left is on the viewer’s right, however, when 3D images are shown that convention is not followed. B: Shows the structural damage to the cortical surface in red (encephalomalacia) where the damage is the sum of what was observed on the T1, T2, and GRE sequences with the yellow representing the white matter signal abnormalities on the FLAIR. C: The axial T1 image has been classified where ventricle is dark black, cortical gray as gray and all subcortical white and gray matter as green, the surface encephalomalacia is highlighted in red. Pediatr Blood Cancer DOI 10.1002/pbc
1726
Bigler
Fig. 3. Another view of how frontal pathology may be shown in 3D, however, in this analysis not only is the cortical surface shown but also all subcortical structures including the putamen (yellow), amygdala (light red), hippocampus (light yellow), thalamus (purple), caudate (dark purple), corpus callosum (gray), and ventricle in light blue. The brainstem is shown in red. Volumes for each of the regions of interest (ROI) can be calculated which may then be examined in relation to neurobehavioral outcome.
be assessed and volumes inferred. This method is referred to as voxel-based morphometry or VBM and can be used to compare and contrast two groups. For example, Figure 4 shows where the greatest loss (anywhere red is shown) of gray matter concentration has occurred in adolescent and young adult individuals who suffered moderate-to-severe TBI and also had a surface contusion [8]. Such VBM techniques can be used to examine the relationship of gray or white matter concentration and cognition [9].
Neuroimaging of White Matter Tracts Diffusion tensor imaging (DTI) is an MRI method that specifically examines the directionality of water diffusion in the brain from which inferences can be made about tissue health and pathway connection. In TBI, a pathway analysis may be extracted from the DTI using a technique referred to as DTI tractography. Figure 5 shows a carefully dissected postmortem brain and corpus callosum tracts compared to a control and a patient who received a TBI. This type of damage often Pediatr Blood Cancer DOI 10.1002/pbc
Fig. 4. A: See-through glass brain depicting areas of decreased gray matter concentration in chronic-stage traumatic brain injury (TBI)— derived from voxel-based morphometry (VBM) in individuals with a history of cortical contusions on the day-of-injury computed tomography (CT) scan. Variations in gray represent the magnitude of significant differences compared to age-matched controls, with the red arrow pointing to the region of greatest concentration in the basal frontal region. B: A different method for depicting the same VBM findings is to plot the differences just on the cortical surface as depicted in red; again showing the distribution of reduced gray matter resulting from TBI. Note the mostly frontotemporal distribution.
relates to cognitive deficits associated with reduced speed of processing [10].
FUNCTIONAL MRI Neuroimaging of the functioning brain can also be achieved with MRI where the MR signal is sensitive to the blood-oxygen dependent level reflected in blood flow. Functional MRI (fMRI) infers regional changes in brain activity as reflecting whether a particular brain region is engaged or not in a time-linked neurobehavioral or neurocognitive task. For example, Figure 6 shows activation within the superior temporal gyrus, which houses auditory cortex, during an auditory task [11]. Additionally, inferences can even be made about functional connectivity in the brain by measuring in-phase features of the blood oxygen-level dependent (BOLD) signal while the brain is at rest—the so-called resting state (rs) fMRI. Also shown in Figure 6, at rest BOLD signal
MRI and Cognitive Function
1727
Fig. 5. The T1 sagittal image on the far left is from a child who sustained a severe traumatic brain injury (TBI) with thinning of the posterior corpus callosum (CC). Evident but CC tractography (mid-left) shows a major loss of tracts throughout the posterior aspect, especially compared to a normal appearing CC in the middle right position. The image on the far right is from a post-mortem dissection that meticulous isolated some of the CC tracts and what their normal appearance should appear like. Used with permission from Gluhbegovic, N and Williams, TH (1980). The human brain: A photographic guide. Philadelphia: Harper & Row.
synchronization is observed in the left and right occipital cortices signifying that a connected network is present between these two areas [12]. This approach permits the examination of brain regions that exhibit functional connectivity. In turn this permits the study of brain networks [12].
Three-Dimensional (3D) Imaging By controlling the magnetic field and frequency of radio-wave pulses, contemporary MRI offers a variety of methods for structural imaging of the brain that comprehensively detects gross brain pathology if present. For example, as demonstrated in Figure 2, the traditional image sequences used in clinical neuroimaging are shown. The T1 image, which is best at identifying overall anatomy, demonstrates some of the structural pathology in the right frontal lobe; but the T2 image is more sensitive in detecting abnormal changes in parenchymal tissue and CSF, actually demonstrating that the damage is even more extensive than detected by just the T1 image. The fluid-attenuated inversion-recovery (FLAIR) sequence is especially sensitive to white matter pathology and various versions of the gradient-recalled echo (GRE) sequence are sensitive in detecting the hemorrhagic by-product hemosiderin, and therefore an excellent technique for detecting prior hemorrhagic lesions. In 2C, the subcortical white and gray matter is isolated from the cortical gray matter and the total amount of cortical damage in the form of residual hemorrhagic abnormalities and cortical wasting (encephalomalacia) shown in read in Figure 2B. The yellow in Figure 2B represents the white matter changes based on the FLAIR image. The tract damage from the loss of white matter integrity in the right frontal lobe shown on the DTI scan shows the actual loss of aggregate fiber tracts involving the frontal lobe.
DISCUSSION Contemporary neuroimaging methods permit excellent visualization of structural damage to the brain that can be quantified and related to neurobehavioral deficits that may arise from injury, infection, developmental factors, disease, and disorder. Pathology may be quantified in a variety of ways which will aid the clinician and researcher to better document outcome in relation to neurobehavioral status.
ACKNOWLEDGMENTS Fig. 6. Top: Auditory activation (blue) indicated by increased blood oxygen-level dependent (BOLD) activation in the left superior temporal lobe gyrus. Bottom: Resting state functional connectivity showing the synchrony of the BOLD signal in homologous brain regions implicating region-specific connectivity (see Refs. [11,12]). Pediatr Blood Cancer DOI 10.1002/pbc
Dr. Bigler affirms that he has no affiliations that he considers to be relevant and important with any organization that to his knowledge has a direct interest, particularly a financial interest, in the subject matter discussed. Such affiliations include, but are not limited to, employment by an industrial concern, ownership of stock, membership on a standing advisory council or committee, a
1728
Bigler
seat on the board of directors, or being publicly associated with a company or its products. The assistance of Jo Ann Petrie, PhD and Tracy Abildskov in the preparation of this manuscript is gratefully acknowledged.
REFERENCES 1. Oldendorf WH. The quest for an image of brain: A brief historical and technical review of brain imaging techniques. Neurology 1978;28:517–533. 2. Bigler ED. Neuroimaging I: Basic science. New York, NY: Plenum Press; 1996. 3. Bigler ED. Hans-Lukas Teuber and ‘The riddle of frontal lobe function in man’ as published in ‘The frontal granular cortex and behavior (1964)’. Neuropsychol Rev 2009;19:9–24. 4. Sullivan EV. Development of brain structures, connections, and functions. Neuropsychol Rev 2010;20:325–326.
Pediatr Blood Cancer DOI 10.1002/pbc
5. FreeSurfer. FreeSurfer is a set of tools for analysis and visualization of structural and functional brain imaging data. FreeSurfer contains a fully automatic structural imaging stream for processing cross sectional and longitudinal data, v. 2013 software. Accessed April 11, 2014: https://surfer.nmr.mgh. harvard.edu 6. Bigler ED, Abildskov TJ, Wilde EA, et al. Diffuse damage in pediatric traumatic brain injury: A comparison of automated versus operator-controlled quantification methods. Neuroimage 2010;50:1017– 1026. 7. Reuter M, Schmansky N, Rosas HD, et al. Within-subject template estimation for unbiased longitudinal image analysis. Neuroimage 2012;61:1402–1418. 8. Bigler ED, McCauley SR, Wu TC, et al. The temporal stem in traumatic brain injury: Preliminary findings. Brain Imaging Behav 2010;4:270–282. 9. Bigler ED. Structural imaging. In: Silver J, McAllister T, Yudofsky S, editors. Textbook of traumatic brain injury. Washington, DC: American Psychiatric Publishing; 2005. pp 73–90. 10. Wu TC, Wilde EA, Bigler ED, et al. Longitudinal changes in the corpus callosum following pediatric traumatic brain injury. Dev Neurosci 2010;32:361–373. 11. Anderson JS, Lange N, Froehlich A, et al. Decreased left posterior insular activity during auditory language in autism. AJNR Am J Neuroradiol 2010;31:131–139. 12. Anderson JS, Druzgal TJ, Froehlich A, et al. Decreased interhemispheric functional connectivity in autism. Cereb Cortex 2011;21:1134–1146.
REVIEW Magnetic Resonance Imaging in the Evaluation of Cognitive Function Erin D. Bigler, Image quality of magnetic resonance imaging (MRI) scans of the brain currently approximate gross anatomy as would be viewed at autopsy. During the first decade of the 21st Century incredible advances in image processing and quantification have occurred permitting more refined methods for studying brain-behaviorcognitive functioning. The current presentation overviews the current status of MRI methods for routine clinical assessment of brain
PhD
1,2,3,4
*
pathology, how these techniques identify neuropathology and how pathological findings are quantified. Diffusion tensor imaging (DTI), functional MRI (fMRI), and resting state fMRI are all reviewed, emphasizing how these techniques permit an examination of brain function and connectivity. General regional relationships of brain function associated with cognitive control will be highlighted. Pediatr Blood Cancer 2014;61:1724–1728. # 2014 Wiley Periodicals, Inc.
Key words: cognitive function; diffusion tensor imaging; functional magnetic resonance imaging; magnetic resonance imaging
INTRODUCTION The role of magnetic resonance imaging in assessing cognitive function in the neurological and/or neuropsychiatric patient has a relatively short history because the ability for in vivo noninvasive imaging of the brain has only been available since the 1970s [1]. Although by the 1980s computed tomography (CT) had become established along with the beginnings of what was then called nuclear magnetic resonance imaging, the forerunner to contemporary magnetic resonance (MR) imaging or MRI, the image quality of the brain was limited as shown in Figure 1. Advancement in image clarity and presentation dominated neuroimaging through the end of the 20th century [2] with ever improving image display as also shown in Figure 1. While in the early days of brain imaging studying brain-cognitive-behavior correlates were problematic because of the coarseness of the images and the limits of how pathology could be identified [3], by the beginning of the 21st century thin-slice MRI approximated what could be viewed at post-mortem as shown in Figure 1. With issues of image quality now achieved, the next advancements occurred in image quantification. Once more refined image quantification was achieved, an explosion in the study of MRI findings and cognition occurred [4].
TISSUE SEGMENTATION At a gross morphological level, three main characteristics of the brain can be identified—white matter, gray matter, and cerebrospinal fluid (CSF) filled spaces and cavities. There are also neuroimaging methods that permit viewing the cerebrovasculature, but in normal brain parenchyma the small vessels and all capillaries embedded within tissue often cannot be differentiated from the white or gray matter wherein they are embedded. The above biological facts provide the basis for a simple classification in neuroimaging where white matter, gray matter, and CSF are delineated in what is referred to as a segmented image as shown in Figure 2C. By identifying a specific color for each tissue type or region of interest (ROI), a color-coded classified image may be generated which in turn provides a method for image quantification since MRI data (signal intensity) used to generate a brain image has a slice thickness and a known distance from the next
C
2014 Wiley Periodicals, Inc. DOI 10.1002/pbc.25110 Published online 11 June 2014 in Wiley Online Library (wileyonlinelibrary.com).
slice. Knowing these metrics provides the basis for volumetric measurements. For example, in Figure 2 from a child who sustained a severe traumatic brain injury (TBI), the different MR sequences were used to isolate the total amount of destructive damage to the cerebral cortex, as shown in red. These areas of damage relate to cognitive deficits in executive and emotional functioning.
TISSUE CLASSIFICATION The next step in image quantification came with classifying anatomical regions. A variety of automated image analysis methods capitalized on key landmarks which provided a roadmap for anatomical identification. Using one type of automated image classification method referred to as FreeSurfer [5] permits the identification and quantification of both cortical and subcortical ROIs. The effectiveness of this technique is that it is reliable and automated making the process ideal for analyzing large numbers of scans within a clinical population compared to a control sample [6,7]. With these types of quantitative measures, relations between certain brain regions and cognition could now be examined [8]. Likewise, pathology can be shown in threedimensional (3D) space as depicted in Figure 3. This child also had a severe TBI with major volume loss involving the left frontal lobe. As in Figure 2, the volume loss is depicted in red, but this time the deficit is in the left frontal area. This type of image presentation can be used to show any type of pathology from degenerative to vascular or neoplastic. But even more importantly, automated programs like FreeSurfer permits the
1
Department of Psychology, Brigham Young University, Provo, Utah; Neuroscience Center, Brigham Young University, Provo, Utah; 3 Department of Psychiatry, University of Utah, Salt Lake City, Utah; 4 The Brain Institute of Utah, University of Utah, Salt Lake City, Utah 2
Correspondence to: Erin D. Bigler, Department of Psychology and Neuroscience Center, 1001 SWKT, Brigham Young University, Provo, UT 84602. E-mail: [email protected] Received 22 April 2014; Accepted 23 April 2014
MRI and Cognitive Function
1725
Fig. 1. Brief history of brain neuroimaging: (A) plain skull film in use since the early 1900s; (B) pneumoencephalography that provided a silhouette of the ventricle after cerebral spinal fluid (CSF) had been replaced with air in use through most of the 1970s; (C) early computed tomography (CT) scan circa 1975; (D) contemporary CT; (E) prototype magnetic resonance imaging (MRI) from the late 1970s; and (F) contemporary 3 Tesla MRI.
volume quantification of all cortical gyri (both white and gray matter along with cortical thickness) as well as all sub-cortical structures. For example, in Figure 3, not only can the cortical gyri be viewed but also the volumes calculated for the hippocampus, amygdala, basal ganglia, ventricular system, midbrain, and brainstem are shown. Once isolated like this, the
potential role of these ROIs in the target outcome behavior can be analyzed. Another approach that capitalizes on the segmented image is to “normalize” each brain within the same 3D space. By using small voxels based on the segmented image that are either white, gray or CSF, a concentration count for each of these tissue types can
Fig. 2. Focal right frontal lesion in traumatic brain injury (TBI) shows differences depending on the magnetic resonance (MR) image sequence. The focal cortical lesions are obvious in the T1 and T2 images (see arrows) but in the fluid-attenuated inversion-recovery (FLAIR) sequence visible extensive white matter differences are evident when comparing the right frontal area (arrow) with the left. The arrow in the gradient-recalled echo (GRE) sequence points to dark, hypointense signal differences that reflect old blood. In the diffusion tensor imaging (DTI) image at this same level note the loss of the distinctiveness of the green colorization which signifies a loss of white matter integrity throughout the right frontal lobe. Green represents pathways that course front to back (anterior to posterior), blue represents vertically oriented tracts and warm colors (orange-red) represents side-to-side coursing tracts. All of these abnormalities can be projected onto a 3D image of the brain as shown in (A). Typically radiological convention is to show the MRI as if looking at the patient so left is on the viewer’s right, however, when 3D images are shown that convention is not followed. B: Shows the structural damage to the cortical surface in red (encephalomalacia) where the damage is the sum of what was observed on the T1, T2, and GRE sequences with the yellow representing the white matter signal abnormalities on the FLAIR. C: The axial T1 image has been classified where ventricle is dark black, cortical gray as gray and all subcortical white and gray matter as green, the surface encephalomalacia is highlighted in red. Pediatr Blood Cancer DOI 10.1002/pbc
1726
Bigler
Fig. 3. Another view of how frontal pathology may be shown in 3D, however, in this analysis not only is the cortical surface shown but also all subcortical structures including the putamen (yellow), amygdala (light red), hippocampus (light yellow), thalamus (purple), caudate (dark purple), corpus callosum (gray), and ventricle in light blue. The brainstem is shown in red. Volumes for each of the regions of interest (ROI) can be calculated which may then be examined in relation to neurobehavioral outcome.
be assessed and volumes inferred. This method is referred to as voxel-based morphometry or VBM and can be used to compare and contrast two groups. For example, Figure 4 shows where the greatest loss (anywhere red is shown) of gray matter concentration has occurred in adolescent and young adult individuals who suffered moderate-to-severe TBI and also had a surface contusion [8]. Such VBM techniques can be used to examine the relationship of gray or white matter concentration and cognition [9].
Neuroimaging of White Matter Tracts Diffusion tensor imaging (DTI) is an MRI method that specifically examines the directionality of water diffusion in the brain from which inferences can be made about tissue health and pathway connection. In TBI, a pathway analysis may be extracted from the DTI using a technique referred to as DTI tractography. Figure 5 shows a carefully dissected postmortem brain and corpus callosum tracts compared to a control and a patient who received a TBI. This type of damage often Pediatr Blood Cancer DOI 10.1002/pbc
Fig. 4. A: See-through glass brain depicting areas of decreased gray matter concentration in chronic-stage traumatic brain injury (TBI)— derived from voxel-based morphometry (VBM) in individuals with a history of cortical contusions on the day-of-injury computed tomography (CT) scan. Variations in gray represent the magnitude of significant differences compared to age-matched controls, with the red arrow pointing to the region of greatest concentration in the basal frontal region. B: A different method for depicting the same VBM findings is to plot the differences just on the cortical surface as depicted in red; again showing the distribution of reduced gray matter resulting from TBI. Note the mostly frontotemporal distribution.
relates to cognitive deficits associated with reduced speed of processing [10].
FUNCTIONAL MRI Neuroimaging of the functioning brain can also be achieved with MRI where the MR signal is sensitive to the blood-oxygen dependent level reflected in blood flow. Functional MRI (fMRI) infers regional changes in brain activity as reflecting whether a particular brain region is engaged or not in a time-linked neurobehavioral or neurocognitive task. For example, Figure 6 shows activation within the superior temporal gyrus, which houses auditory cortex, during an auditory task [11]. Additionally, inferences can even be made about functional connectivity in the brain by measuring in-phase features of the blood oxygen-level dependent (BOLD) signal while the brain is at rest—the so-called resting state (rs) fMRI. Also shown in Figure 6, at rest BOLD signal
MRI and Cognitive Function
1727
Fig. 5. The T1 sagittal image on the far left is from a child who sustained a severe traumatic brain injury (TBI) with thinning of the posterior corpus callosum (CC). Evident but CC tractography (mid-left) shows a major loss of tracts throughout the posterior aspect, especially compared to a normal appearing CC in the middle right position. The image on the far right is from a post-mortem dissection that meticulous isolated some of the CC tracts and what their normal appearance should appear like. Used with permission from Gluhbegovic, N and Williams, TH (1980). The human brain: A photographic guide. Philadelphia: Harper & Row.
synchronization is observed in the left and right occipital cortices signifying that a connected network is present between these two areas [12]. This approach permits the examination of brain regions that exhibit functional connectivity. In turn this permits the study of brain networks [12].
Three-Dimensional (3D) Imaging By controlling the magnetic field and frequency of radio-wave pulses, contemporary MRI offers a variety of methods for structural imaging of the brain that comprehensively detects gross brain pathology if present. For example, as demonstrated in Figure 2, the traditional image sequences used in clinical neuroimaging are shown. The T1 image, which is best at identifying overall anatomy, demonstrates some of the structural pathology in the right frontal lobe; but the T2 image is more sensitive in detecting abnormal changes in parenchymal tissue and CSF, actually demonstrating that the damage is even more extensive than detected by just the T1 image. The fluid-attenuated inversion-recovery (FLAIR) sequence is especially sensitive to white matter pathology and various versions of the gradient-recalled echo (GRE) sequence are sensitive in detecting the hemorrhagic by-product hemosiderin, and therefore an excellent technique for detecting prior hemorrhagic lesions. In 2C, the subcortical white and gray matter is isolated from the cortical gray matter and the total amount of cortical damage in the form of residual hemorrhagic abnormalities and cortical wasting (encephalomalacia) shown in read in Figure 2B. The yellow in Figure 2B represents the white matter changes based on the FLAIR image. The tract damage from the loss of white matter integrity in the right frontal lobe shown on the DTI scan shows the actual loss of aggregate fiber tracts involving the frontal lobe.
DISCUSSION Contemporary neuroimaging methods permit excellent visualization of structural damage to the brain that can be quantified and related to neurobehavioral deficits that may arise from injury, infection, developmental factors, disease, and disorder. Pathology may be quantified in a variety of ways which will aid the clinician and researcher to better document outcome in relation to neurobehavioral status.
ACKNOWLEDGMENTS Fig. 6. Top: Auditory activation (blue) indicated by increased blood oxygen-level dependent (BOLD) activation in the left superior temporal lobe gyrus. Bottom: Resting state functional connectivity showing the synchrony of the BOLD signal in homologous brain regions implicating region-specific connectivity (see Refs. [11,12]). Pediatr Blood Cancer DOI 10.1002/pbc
Dr. Bigler affirms that he has no affiliations that he considers to be relevant and important with any organization that to his knowledge has a direct interest, particularly a financial interest, in the subject matter discussed. Such affiliations include, but are not limited to, employment by an industrial concern, ownership of stock, membership on a standing advisory council or committee, a
1728
Bigler
seat on the board of directors, or being publicly associated with a company or its products. The assistance of Jo Ann Petrie, PhD and Tracy Abildskov in the preparation of this manuscript is gratefully acknowledged.
REFERENCES 1. Oldendorf WH. The quest for an image of brain: A brief historical and technical review of brain imaging techniques. Neurology 1978;28:517–533. 2. Bigler ED. Neuroimaging I: Basic science. New York, NY: Plenum Press; 1996. 3. Bigler ED. Hans-Lukas Teuber and ‘The riddle of frontal lobe function in man’ as published in ‘The frontal granular cortex and behavior (1964)’. Neuropsychol Rev 2009;19:9–24. 4. Sullivan EV. Development of brain structures, connections, and functions. Neuropsychol Rev 2010;20:325–326.
Pediatr Blood Cancer DOI 10.1002/pbc
5. FreeSurfer. FreeSurfer is a set of tools for analysis and visualization of structural and functional brain imaging data. FreeSurfer contains a fully automatic structural imaging stream for processing cross sectional and longitudinal data, v. 2013 software. Accessed April 11, 2014: https://surfer.nmr.mgh. harvard.edu 6. Bigler ED, Abildskov TJ, Wilde EA, et al. Diffuse damage in pediatric traumatic brain injury: A comparison of automated versus operator-controlled quantification methods. Neuroimage 2010;50:1017– 1026. 7. Reuter M, Schmansky N, Rosas HD, et al. Within-subject template estimation for unbiased longitudinal image analysis. Neuroimage 2012;61:1402–1418. 8. Bigler ED, McCauley SR, Wu TC, et al. The temporal stem in traumatic brain injury: Preliminary findings. Brain Imaging Behav 2010;4:270–282. 9. Bigler ED. Structural imaging. In: Silver J, McAllister T, Yudofsky S, editors. Textbook of traumatic brain injury. Washington, DC: American Psychiatric Publishing; 2005. pp 73–90. 10. Wu TC, Wilde EA, Bigler ED, et al. Longitudinal changes in the corpus callosum following pediatric traumatic brain injury. Dev Neurosci 2010;32:361–373. 11. Anderson JS, Lange N, Froehlich A, et al. Decreased left posterior insular activity during auditory language in autism. AJNR Am J Neuroradiol 2010;31:131–139. 12. Anderson JS, Druzgal TJ, Froehlich A, et al. Decreased interhemispheric functional connectivity in autism. Cereb Cortex 2011;21:1134–1146.
