Main Content Region

Analysing MRI data

Colm G. Connollya, Erik O’Hanlona, Liam Nestora, Jay Nierenbergb, Marina Shpanerb, John J. Foxeb,c Hugh Garavana
a School of Psychology and Institute of Neuroscience, The Lloyd Building, University of Dublin, Trinity College, Dublin 2, Éire.
b Department of Psychology, Program in Cognitive Neuroscience, The City College of the City University of New York, North Academic Complex, 138th Street and Convent Avenue, New York, NY 10031, USA
c Cognitive Neurophysiology Laboratory, Nathan S. Kline Institute for Psychiatric Research, Cognitive Neuroscience and Schizophrenia Program, 140 Old Orangeburg Road, Orangeburg, NY 10962, USA

Here at the Trinity College Institute of Neuroscience we are in the privilaged position to have Ireland’s only research dedicated 3 Telsa whole body MRI scanner. Its recent introduction has provided an opportunity to pursue exciting and novel research questions using cutting edge technologies.

This allows us not only to record brain activity as it unfolds over time but also to track the white matter fibres of the brain and to determine relative volumes of the grey and white matter in the brain. The grey matter is the thin layer of the brain on the surface which is, by and large, responsible for our ability to talk, see hear and touch, among other functions. The white matter is akin to the information super highway of the brain as it connects various grey matter regions with one another and other sub-cortical brain regions.

The ability to observe brain activity over time is based on the fact that highly active neurons consume more oxygen than those that are less active. The oxygen rich blood that is sent to the more active regions of the brain has slightly different magnetic properties to blood that has given up its oxygen to the surrounding tissue. This difference gives rise to the blood oxygenation level–dependent (BOLD) signal and can be recorded with magnetic resonance imaging.

Typically the analysis of MRI images is computationally very intensive and we therefore benefit greatly from the resources available in TCHPC. Some studies can take a few hours to analyse, others can take a week or more of continuous compute time.

One of the studies that we have analysed using the resources provided to us by TCHPC is a study to determine the effect of abstinence in cocaine abusers. Cocaine is one of the most powerful reinforcing substances known to man (Kuhar, Ritz, & Boja, 1991). Neuroimaging studies have revealed that metabolic differences are present in anterior cingulate cortex (ACC) and lateral prefrontal cortex (LPFC) relative to non-cocaine abusing people (Stapleton et al., 1995; Volkow, Mullani, Gould, Adler, & Krajewski, 1988). In this study we attempted to determine whether the amount of activity in brain regions typically associated with response inhibition tasks varied depending on the length of time cocaine users abstain from consuming the drug. Figure 1 depicts some of the results from this study. We observed many regions of the brain that showed statistically significant differences including those usually associated with the type of response inhibition task used in this experiment. Figure 1 shows some of these clusters for erroneous responses, that is responses which should not have been made during the course of the experiment.

Figure 1 The above figure depicts the regions of the brain identified as significantly different between the brains of cocaine addicts and control participants for errors of commission, that is, erroneous responses that should not have been made during the performance of the task. The green cross hairs depict the extent of the cutting planes used in the creation of the 3D image in the bottom right corner. The cross hairs are centred on the anterior cingulate, a region of the cortex thought to play a crucial role in the type of response inhibition task used in this experiment.

Another study that has been conducted in TCIN relates to the effect of cannabis consumption on human learning and memory, impairments of which have been linked to cannabis use. Cannabis-related impairments in learning and memory in chronic cannabis users, it has been argued, are caused by the effects of cannabis on hippocampal functioning. This study used a task in which subjects had to learn the names of numerous faces, to examine cortical and (para)hippocampal activity during learning and recall in 14 current users of cannabis and 14 controls. Results showed that despite non-significant differences in learning and memory performance, cannabis users had significantly lower levels of BOLD activity in the right superior temporal gyrus, right superior frontal gyrus, right middle frontal gyrus and left superior frontal gyrus compared to controls during learning. Results also showed that cannabis users had significantly higher BOLD activity in the right parahippocampal gyrus during learning. Hypoactivity in frontal and temporal cortices, and relative hyperactivity in the parahippocampus identify functional deficits and compensatory processes in cannabis users.

This study also investigated whether there were any observable changes in the white matter tracts, using Diffusion Tensor imaging (DTI), of the cannabis users relative to the non-drug using participants.

Diffusion-weighted imaging is an MR imaging technique that relies on the motion of water molecules in the brain. Ideally, water molecules will diffuse evenly in all directions (isotropic movement) when unrestricted. In practice, this motion is limited by the complex tissue composition of the brain that creates barriers and restricts the motion causing the water molecules to follow paths of least resistance. Diffusion tensor imaging relies on this uneven movement pattern (anisotropic diffusion) and quantifies the principle direction of the motion. The macroscopic diffusion properties are sensitive to tissue structure and hence offer a mechanism to indirectly interrogate on a microscopic level, i.e. at a cellular level. The measured diffusion tensor information is used to calculate measures such as Fractional anisotropy (FA) or the degree of asymmetric water diffusion, which can be used to assess the integrity of the white matter tissue. Thus FA can be used to examine white matter tissue differences between the various cohorts of interest (see figure 2). Moreover, the principal diffusion direction tensors can be used to infer the white matter fibre tracts or paths within the tissue.

Figure 2(a) below shows a sample diffusion tensor image from this study. The colours correspond to the principal fiber directions. Red fibers run left to right; blue run anterior to posterior; and green run superior to inferior. On the right in figure 2(b) an average white matter skeleton for all the subjects in the study is shown in green. The red blob is a region of white matter that differs between the cannabis using participants and the non-using controls. Figure 3 shows a tract of white fibers that was identified as connecting the right occipital lobe to the right superior frontal gyrus.

Figure 2 (a) Diffusion Tensor image (DTI) modulated by its fractional anisotropy. (b) Localized FA group difference (shown in red) with a mean white matter fibre-tract skeleton representation shown in green.

Figure 3 (a) An example of a white matter fibre tract (shown in red / yellow) superimposed on a background image of mean white matter FA in the sagittal plane. This demonstrates a pathway between the occipital lobe and the superior frontal gyrus (show in blue) in a coronal slice of the brain demonstrating white matter connective pathway between the parahippocampal gyrus and the superior frontal gyrus.

The knowledge gained from these experiments has the potential to allow us to understand how the brain changes in response to substances, such as cocaine or cannabis, or responds to the illnesses such as schizophrenia. These changes may be functional, that is different regions may activate more or less for a particular task. Or they may be structural, such as changes in regional grey matter concentration or changes in the integrity of white matter tracts.


  1. Kuhar, M. J., Ritz, M. C., & Boja, J. W. (1991). The dopamine hypothesis of the reinforcing properties of cocaine. Trends in Neurosciences, 14(7), 299-302.
  2. Stapleton, J. M., Morgan, M. J., Phillips, R. L., Wong, D. F., Babington, C. K., Yung, M. D., et al. (1995). CEREBRAL GLUCOSE-UTILIZATION IN POLYSUBSTANCE ABUSE. Neuropsychopharmacology, 13(1), 21-31.
  3. Volkow, N. D., Mullani, N., Gould, K. L., Adler, S., & Krajewski, K. (1988). CEREBRAL BLOOD-FLOW IN CHRONIC COCAINE USERS - A STUDY WITH POSITRON EMISSION TOMOGRAPHY. British Journal of Psychiatry, 152, 641-648.