Valorie N. Salimpoor
Brain imaging technology has rapidly advanced in the last three decades. Improvements in structural brain imaging techniques in the 1970s were followed by development of sophisticated functional neuroimaging techniques in the 1980's, and have significantly improved in the past decade. Of these neuroimaging techniques, MRI, fMRI, EEG, ERP, and MEG are uniquely suited to study structural, physiological, and developmental brain abnormalities in children and to perform repeated measures because they involve no ionizing radiation or radioactive isotopes, and they have been shown to have an absence of biological hazards at currently used field strengths. This paper will briefly introduce current brain imaging techniques used to diagnose and study developmental disabilities and discuss their clinical and research applications, as well as advantages and disadvantages. Their applications will be discussed with respect to Autism Spectrum Disorder (ASD).
Brain imaging technology has rapidly advanced in the last three decades. Structural neuroimaging techniques such as computed tomography (CT) in the 1970s quickly gave way to magnetic resonance imaging (MRI) in the 1980s, followed by sophisticated functional neuroimaging techniques developed later that decade. Presently, there are two basic types of neuroimaging techniques. The first, structural brain imaging, provides static images of the brain at rest. This technique allows for localization of lesions or damage in the brain, measurement and comparison of different brain regions, and can be used to monitor changes over time in areas that are being impacted by some kind of damage. Examples of structural neuroimaging includes CT scans and MRI. The second type of brain imaging, functional neuroimaging, measures brain activity during the performance of a task and allows researchers to determine which parts of the brain are activated at that time. Examples of functional neuroimaging techniques include functional MRI (fMRI), positron emission tomography (PET) scans, single photon emission computerized tomography (SPECT), magnetoencephalography (MEG), event-related potentials (ERP), and electroencephalograms (EEG). Structural and function methods are both important for unveiling the neurological correlates of developmental disabilities.
This paper will review the different structural and functional neuroimaging techniques that are currently used in the assessment and study of developmental disabilities. The clinical applications, implications for research, and benefits and side-effects of each technique will be presented, with respect to Autism Spectrum Disorder (ASD). Autism impacts nearly every aspect of a child's perception of and interaction with the surrounding world and has complex effects on many interacting brain systems. Exploring the neurobiology of ASD is a crucial step in the understanding the symptoms and the development of assessment, diagnosis, intervention, and treatment programs.
Computerized Tomography (CT)
Computerized tomography, developed in the 1970's, is known for being the first widely used imagery technique to reveal brain anatomy. By combining X-ray techniques with computerized data acquisition and image reconstruction, CT provides 15-20 horizontal-sliced images of the brain. This type of imagery presents a picture of the human brain by passing x-ray beams through the head at various angles. Different parts of the brain have different densities and are distinguished by their different absorptions of x-rays. Dense tissue such as bone appears white and material with the least density, such as cerebrospinal fluid, appears black. Although CT has poor resolution and does not show the details of the brain to the extent that is now possible with other techniques, the advent of CT was a great medical advancement because of its ability to detect tumors, lesions, and other brain abnormalities.
Magnetic Resonance Imaging (MRI)
Over the past 20 years, innovations in physics and computer science have promoted MRI as an essential tool for investigating the biological substrates of developmental disabilities. Contrary to CT scans this techniques requires no radiation exposure, and can be used for exploratory studies involving a large number of slices in any single patient. MRI is now one of the preferred imaging technique for pediatric populations (Eliez & Reiss, 2000). Since MRI can be performed on healthy individuals who can then also participate in behavioral testing, this permits investigations of whether individual differences in specific cognitive abilities or patterns of hemispheric specialization can have measurable anatomical correlates. Furthermore, MRI can produce very clear and detailed pictures of brain structures.
MR technology works with the electromagnetic phenomenon in biological tissues to generate images. The physics of MRI is complicated and will be discussed briefly. First, the atom's nuclei are brought into alignment by a strong magnetic field. A brief pulse of radio waves is then introduced, which dislodges the spinning hydrogen nuclei. Each active nucleus at a given magnetic field strength has a unique radio frequency. Following the pulse, the nuclei return to alignment with the magnetic field, and emit radio waves themselves in the process, which are detected by antennae-like receivers situated around the head. White and grey matter, cerebrospinal fluid, blood, and other tissue emit different radio waves. This difference can be use to construct images of the anatomy of the brain.
MRI provides clear and life-like detailed views of the brain. Furthermore, whereas CT requires the use of invasive x-rays, MRI acquires images non-invasively through the use of a strong magnetic field and radio waves. Like CT, MRI affords the possibility of studying relationships between localized brain injuries or abnormalities and the type of cognitive deficits that ensue. Unlike CT, MRI has a great facility to obtain images of the brain at different angles or planes of view, as well as the possibility of constructing three-dimensional representations of regions of interest. Since MRI can be performed on healthy individuals who can then also participate in behavioral testing, this permits investigations of whether individual differences in specific cognitive abilities or patterns of hemispheric specialization can have measurable anatomical correlates.
Functional Magnetic Resonance Imaging (fMRI)
Recent advances in MRI technology have resulted in functional or high-speed MRI, which can reveal brain activity in addition to brain structure. This type of imaging can reveal brain activity through the measurement of changes in blood flow, blood volume, and blood oxygenation (e.g., Belliveau et al., 1991). When neurons become more active, they increase their need for oxygen, which results in a regional increase in blood flow and a corresponding increase in the conversion of blood oxyhemoglobin to deoxyhemoglobin. Water molecules produce different radio signals in the presence of oxyhemoglobin and deoxyhemoglobin because these two forms of hemoglobin differ in their magnetic parameters to be sensitive to these effects, resulting in a physiological index of neuronal activity. Presently, this technique has a temporal resolution of about one second for discerning physiological changes, being superior to that of PET but not as sensitive as EEG and MEG. Because functional MRI can be performed noninvasively, individual subjects can participate in an unlimited number of sessions. The future of fMRI is extremely promising, especially given the current wide-spread use of MRI technology, which in many instances can be adapted to include fMRI capabilities and relate structure to function.
Positron Emission Tomography (PET)
PET creates a visual image of functioning in various parts of the brain by tracing chemical activity. Similar to fMRI, PET works by measuring cerebral blood flow and the consumption of oxygen and glucose, the two virtually exclusive substrates of cerebral energy metabolism. However, unlike fMRI, PET is invasive with brief dosage of ionizing radiation, and is sensitive to factors such as anxiety and distraction, making the results difficult to interpret. A short-lived radio isotope is labeled to become a positron-emitting isotope and a small quantity is then injected into the body. The molecule is unstable but reaches a more stable non-radioactive state by releasing a charged particle. The energy created from the release of the positron as it courses through the body is used to produce the PET image. Brain cells use glucose as fuel, and PET works on the theory that if brain cells are more active, they will consume more of the radioactive glucose, and if less active, they will consume less of it. A computer uses the absorption data to show the levels of activity as a color-coded brain map, with one color (usually red) indicating more active brain areas, and another color (usually blue) indicating the less active areas. PET imaging software allows researchers to look at cross-sectional slices of the brain, and therefore observe deep brain structures, which earlier techniques like EEGs could not. PET allows direct assessment of brain biochemistry and is currently one of the most sophisticated and expensive of the functional imaging techniques. However, PET and SPECT (discussed below) are not often used on children. Between the two, SPECT is the preferred method for pediatric populations. However, the use of a radioactive tracer is rarely approved for the use in healthy children, and comparisons with control groups are difficult for research purposes.
Single Photon Emission Computed Tomography (SPECT)
Functional images of the brain can also be acquired by techniques that measure regional cerebral blood flow through the use of the more accessible single photon-emitting isotopes. This approach has, in turn, been combined with CT technology resulting in SPECT. Although SPECT is most commonly used to measure cerebral blood flow, it is also being applied to neurotransmitter studies. Whereas SPECT is considerably less costly than PET, the trade-off is in less detailed images and physiological information that is less amenable to quantification. Nevertheless, SPECT is proving to be a valued instrument in clinical and research settings. For instance, in the study of epilepsy, SPECT can be useful in localizing brain regions associated with seizure generation (Harvey & Berkovic, 1994). The neurosurgical removal of such abnormal brain areas is a possible treatment for patients with severe epilepsy who are unresponsive to drug therapy. Both blood flow and PET studies can even reveal regional brain activity that occurs during different types of thinking, in the absence of sensory stimulation or motor responses.
Electroencephalograms (EEG) and Event-Related Potentials (ERP)
EEG, was one of the first - and still very useful - ways of non-invasively observing human brain activity. An EEG is a recording of electrical signals from the brain made by hooking up electrodes to the scalp. These electrodes pick up electric signals naturally produced by the brain and send them to galvanometers, which detect and measure small electric currents, and produce graphical images of brain activity. EEG has excellent temporal resolution and continues to be the primary clinical tool for studying epilepsy. Epileptic patients who are candidates for neurosurgical treatment sometimes have recording electrodes surgically placed directly onto the cortical surface or inserted into the depths of the brain, allowing for more accurate localization of areas that generate seizures. EEGs can be used to create a moving picture of the regional changes in brain activity when subjects perform a particular task. This procedure is called event-related potential (ERP). Recordings can be acquired from patients while they perform perceptual, motor, or cognitive tasks. A possible task could be presenting successive series of familiar and unfamiliar photographs of faces, and for each photograph patients respond to as to whether they recognize the face. The ERP technique would reveal the precise timing and approximate location of brain regions with recordable activity at each stage of processing: from visual input, through various cognitive stages, to response out put. This method is often used with children because it is easy to apply, inexpensive, and invasive.
MEG is a technique that measures biomagnetic information in the brain. All electrical currents generate a magnetic flux, thus it follows that currents produced by neurons also produce magnetic fields. With MEG, the magnetic flux generated within the brain is detected by superconducting antenna coils connected to extremely sensitive amplifiers known as superconducting quantum interference devices (SQIDs). For EEG, recording electrodes are attached to the scalp, whereas for MEG the antenna coils are permanently positioned within a captor that can be simply lowered over the patients head. One advantage of measuring magnetic fields is that, unlike electrical signals, magnetic fields pass easily through tissue and bone without being distorted. Presently, MEG, like PET, is extremely expensive and its accessibility is limited to research studies.
Neuroimaging and Autism Spectrum Disorder
Structural neuroimaging techniques such as CT and MRI, have revealed important information about the neurobiological basis of ASD. These techniques can locate anatomical abnormalities. For examples, CT and MRI studies have found enlargement of the right lateral ventricle (McKelvey, et al., 1995), which implies loss of tissue in the right hemisphere, and areas of abnormal gyration bilaterally (Berthier, Starkstein, & Leguarda, 1990), which implies anomalies in cortical migration during brain development. Another study found damage to the left temporal lobe (Jones & Kerwin, 1990)-an area responsible for emotions and recognition of facial expressions, both of which are implicated in ASD. MRI studies of ASD have found abnormality of the dorsolateral prefrontal cortex and the left temporal lobe near the amygdala (Volkmar et al., 1996)-two regions responsible for Theory of Mind (ToM), a task with which individuals with ASD have difficulty. This task requires individuals to infer the feelings and intentions of others. Other studies have found abnormalities near the emotional centre of the brain -the limbic system (Nieminen-von Wendt et al., 2002), and two regions responsible for motor control and movement (Berthier et al., 2003; McAlonan et al., 2002). These symptoms have all been implicated in ASD.
Although structural techniques provide information about which areas may be damaged, they cannot provide information relating brain structure to function. On the other hand, functional brain imaging techniques provide information about which structures are required to perform specific functions, and how these pathways may be deficient or disrupted. An fMRI study looked at regional changes in blood flow during a social attribution task, which required the assessment of social motives and intentions (Schultz et al., under review). Normally, during performance of such tasks the medial prefrontal cortex, an area known to be involved in ToM tasks, is activated (Oktem, Kiren, Karaagaoglu, & Anlar, 2001). However, Schultz et al.'s study found bilateral activation of the angular gyrus. These results suggest that children with ASD may be using different neural pathways to complete these tasks. PET studies have also looked at ToM task performance in individuals with ASD. Similar to fMRI studies, PET studies have found altered frontal lobe activation. As mentioned above, ToM tasks activate the left medial prefrontal cortex. However, this activation was absent in individuals with ASD. Instead, activity was observed in immediately adjacent areas.
In summary, structural and functional neuroimaging techniques are both important for unveiling the etiology, course of disability, and treatment and rehabilitation outcomes. Of the techniques presented, MRI, fMRI, EEG, ERP and MEG are uniquely suited to study structural, physiological, and developmental brain abnormalities in children and to perform repeated measures because they are not invasive and involve no ionizing radiation or radioactive isotopes.
Recent advances in neuroimaging offer the possibility to understand more about the etiology of developmental disabilities allowing for the development of effective treatment. Furthermore, they provide means of measuring the validity and progress of rehabilitation programs.
Belliveau, J. W., Kennedy, D. N., McKinstry, R. C., Buchbinder, B. R., Weisskoff, R.M., Cohen, M. S., Vevea, J. M., Brady, T. J., & Rosen, B. R. (1991). Functional mapping of the human visual cortex by magnetic resonance imaging. Science, 254, 716-719.
Berthier, M. L., Starkstein, S. E., & Keiguarda, R. (1990). Developmental cortical anomalies in Asperger's syndrome: neuroradiological findings in two patients. Journal of Neuropsychiatry & Clinical Neurosciences, 2, 197-201.
Berthier, R., Rizzitelli, A., Martinon-Ego, C., Laharie, A. M., Collin, V., Chesne, S., & Marche, P. N. (2003). Comorbid Asperger and Tourette syndromes with localized mesencephalic, infrathalamic, thalamic, and striatal damage. Developmental Medicine & Child Neurology, 45, 207-212.
Eliez, S., & Reiss, A. L. (2000). MRI neuroimaging of childhood psychiatric disorders: a selective review. Journal of Child Psychology & Psychiatry & Allied Disciplines, 41, 679-694.
Harvey, A., S., & Berkovic, S. F. (1994). Functional neuroimaging with SPECT in children with partial epilepsy. Journal of Child Neurology, 9(Suppl 1), S71-81.
Jones, P. B., & Kerwin, R. W. (1990). Left temporal lobe damage in Asperger's Syndrome. British Journal of Psychiatry, 156, 570-572.
McAlonan, G. M., Daly, E., Kumari, V., Critchley, H. D., can Amerlsvoort, T., Suckling, J., Simmons, A., Sigmundsson, T., Greenwood, K., Russeell, A., Schmitz, N., Happe, F., Howlin, P., & Murphy, D. G. (2002). Brain anatomy and sensorimotor gating in Asperger's syndrome. Brain, 125, 1594-1606.
McKelvey, J. R., Lambert, R., Mottron, L., & Shevell, M. I.. Right-hemisphere dysfunction in Asperger's syndrome. Journal of Child Neurology, 10, 310-314.
Nieminen-von Wendt, T., Salonen, O., Vanhala, R., Kulomaki, T., von Wendt, L., & Autti, T. (2002). A quantitative controlled MRI study of the brain in 28 persons with Asperger syndrome. International Journal of Circumpolar Health, 61, 22-35.
Oktem, F., Diren, B., Karaagaoglu, E., & Anlar, B. (2001). Functional magnetic resonance imaging in children with Asperger's syndrome. Journal of Child Neurology, 16, 253-256.
Schultz, R. T., Klin, A., van der Gaag, C., Skudlarski, P., Herrington, J. & Gore, J. C. Medial prefrontal involvement in the process of social attribution: An fMRI study. Under review.
Volkmar, F. R., Klin, A., Schultz, R., Bronen, R., Marans, W., Sparrow, S., & Cohen, D. J. (1996). Asperger syndrome. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 118-123.
For further information:
257 Madison Heights Blvd., Markham, ON, L6C-2L9, Canada