Imaging in Neurosciences & Mental Health

Brain diseases represent a leading cause of disability, dementia, and health-care utilization in Canada. Some brain illnesses such as epilepsy and schizophrenia begin early and last a lifetime. Others, such as Alzheimer disease occur late in life and are ultimately fatal. Imaging the brain by MRI provides a glimpse into the structure, function, and metabolism of the brain without the use of potentially harmful radiation. In recent years, the development of new MRI methods has sparked enormous excitement in the medical community. BIRC scientists are leading new programs to develop high magnetic field imaging methods that promise to provide greater image resolution, greater sensitivity, and the ability to visualize disease processes using new methods.

Understanding the Normal Brain

A first step in understanding brain disease is to understand normal brain structure, function, and biochemistry. BIRC researchers are using state-of-the-art magnetic resonance imaging scanners including two 3-Tesla MRIs, the only human 7-Tesla MRI in Canada, and the only 9.4-Tesla medium bore MRI in Canada, to study all aspects of the brain. Some of this work is done in collaboration with scientists from the Brain and Mind Institute at Western University. Fundamental discoveries in how the brain works fuel these efforts to study brain disease such as Alzheimer Disease, Multiple Sclerosis, and Epilepsy.

Dementia (Frontotemporal and Alzheimer Disease)

These research programs focus on the use of neuroimaging techniques to delineate early dysfunction in the dementias, to develop improved tools for early diagnosis, and determine the functional mechanisms underlying patients’ symptoms. Highly innovative paradigms for symptom-targeted cognitive tasks with fMRI (Finger) are being combined with serologic biomarkers in frontotemporal dementia patients. The use of high resolution MRI to identify the brain regions that atrophy in prodromal Alzheimer disease, the use of highly localized magnetic resonance spectroscopy methods (i.e. ≤1 cc voxels at 7T) to define the metabolic changes associated with cognitive decline (Bartha) and response to cholinergic treatment in Alzheimer Disease (Bartha, Borrie), and the use of Magnetic Resonance Spectroscopy (MRS) and volumetric imaging to understand the neural substrates of cognitive decline associated with increased gait variability (Bartha, Montero-Odasso) are other major thrusts in this research area.

Multiple Sclerosis (MS)

MS is the most common chronic neurological disease affecting young adults, and of particular relevance because Canada’s MS prevalence rate is amongst the highest in the world. Conventional diagnostic MRI findings have not correlated well with clinical symptoms and have not demonstrated significant predictive power for disease progression. This is likely because conventional MRI visualizes only a few of the lesions and damage seen in histopathological studies. The Menon group has developed a high-resolution version of their multi-echo RASTAMAP sequence for measuring the intrinsic tissue magnetic parameters R2*, the microscopic magnetic fields in tissue (mB0) and the tissue magnetic susceptibility (c), all of which can be linked to histopathological findings in MS (and likely many other diseases such as epilepsy and PD as described below).

The team is pursuing this at 3T and 7T. The 3T data shows lesions that are homogeneous on conventional MRI, but can be stratified using this quantitative sequence. However, as it is a susceptibility-weighted sequence, motion needs to be carefully controlled. At 7T, there are strong indications of grey matter involvement in the disease from the MGH group and from data at this centre.

Epilepsy – Pinpointing the Focus

The goal of the research program in Epilepsy imaging is to identify where seizures begin in the brain.  Using a combination of 3 Tesla and 7 Tesla MRI, researchers (Peters, Burneo, Bartha) are examining the brains of people with seizures originating in the temporal lobes, but which are difficult to see using conventional clinical imaging. Using advanced techniques, including high-resolution imaging, magnetic resonance spectroscopy to measure metabolism, diffusion tensor imaging to examine microstructure, and functional MRI (Figure below) researchers hope to uncover a pattern that is indicative of seizure activity. This work is combined with ultra-high resolution imaging of excised epileptogenic tissue from people who have undergone brain surgery, to understand how the brain changes from seizures. These methods will increase the ability of surgeons to ascertain the focal point of a seizure so this tissue can be effectively removed while sparing normal tissue.

Temporal lobe epilepsy is the most common form of medically-intractable focal epilepsy, and is commonly treated with standard surgical resection of the anterior temporal lobe.  Unfortunately only 60% of patients become seizure free following surgery, and it is thought that improvements in refining the extent can lead to better surgical outcomes and potentially lead to less invasive surgical approaches. We are investigating the use of quantitative multi-parametric MRI to more accurately localize the site of seizure onset and guide surgical resection of identified lesions. Our unique study protocol involves 3T and 7T imaging of patients prior to surgery, MR microscopy of resected surgical tissue samples, and histological processing of this tissue to investigate histopathological correlations with imaging.  We have developed novel methodology to non-linearly warp digital histology slides to the pre-operative images to allow for direct comparison of quantitative imaging parameters with pathology. 

This will allow us to determine the quantitative MRI signature of pathological tissue, thereby enabling the pre-operative prediction of pathology in-vivo.  We are using machine learning techniques, such as support vector machines, to build patient-specific classifiers that can lateralize and localize imaging abnormalities and relate these to histopathological findings.  This will work will ultimately translate into novel surgical planning and image-guided technology to aid neurosurgeons towards less invasive and more precise resections in epilepsy surgery.

Illustration of MRI and pathology pipeline for investigating quantitative MRI signatures of histopathology in temporal lobe epilepsy

Spinal Cord Compression

In its mildest form, cervical spondylosis can cause symptoms such as stiffness, restricted range of motion as well as neck and arm pain. Surgical outcomes are unpredictable, likely depending on whether the spinal cord has undergone reversible or irreversible injury. By applying fMRI and MRS at 3T in a CIHR funded study, members of this team (Duggal, Bartha) have recently demonstrated cortical reorganization and recruitment of surrounding brain cortex to perform a motor task following surgery. Brain metabolite levels were altered in patients with reversible spinal cord compression as well. The future of this program involves studying the spine directly to refine the criteria used to choose surgical candidates. However, studies to examine the spine with DTI and MRS are highly susceptible to motion and will require development of faster imaging and specific shim coils to improve magnetic field homogeneity within the cord.

Parkinson’s Disease (PD)

The effect of dopamine depletion on the neurochemical pathways linking the basal ganglia, substantia nigra (SN), thalamus and cerebral cortex are not well understood. Tremor dominant (TD), akinetic rigid (AkR), and primary instability of gait (PIGD) are 3 clinical subtypes of PD. These subtypes are associated with separate patterns of change in the neurochemical pathways involved in PD. The objective of the work lead by Dr. Jog is to measure the neurometabolic profile within the SN, subthalamic nucleus (STN) and ventral anterior- ventrolateral thalamus (VA/VL) in patients with these 3 clinical subtypes and in healthy control subjects using short echo time magnetic resonance spectroscopy. In additional to their motor symptoms, 15-20% of PD patients suffer from dementia (x6 more likely than controls), or a less severe cognitive impairment that is an important predictor of quality of life. The team is building on Owen’s substantial existing body of work on the neurochemical and neuroanatomical basis of cognitive deficits in PD and their implications for frontostriatal dysfunction. Longitudinal imaging studies being developed will investigate the cognitive, anatomical and genotypic mechanisms determining the transition from PD to dementia within a large cohort of patients. Cross-sectional studies within sub-populations or individuals will examine the effects of medication, genotype and cognitive training. To do this, they are developing a common battery of fMRI paradigms that produce specific, performance-related activity in key cortical and sub-cortical regions (e.g. frontal-lobe, basal-ganglia and hippocampus) and obtain functional and structural connectivity measures in patients and volunteers to investigate how these regions functionally interact.

Acute Brain Injury

Improvements in intensive care have led to an increase in the number of patients who survive severe brain injury. Some have a good prognosis, but many progress to a vegetative (VS) or minimally conscious state (MCS). Recent advances in fMRI, led by Dr. Owen and his team, have shown that fMRI can be used to assess cognitive functions in VS and MCS without the need for any overt response from the patient. Up to 20% of behaviourally vegetative patients may have some level of preserved awareness leading other key scientists in this field to conclude that integrating the fMRI techniques pioneered by Adrian Owen and his team with existing clinical and behavioural assessments will be essential to reduce diagnostic error in these patients. These diagnostics have critically important medical, legal and ethical implications.

Owen is extending his landmark studies of acute brain injury at the Centre for Functional Brain Mapping within BIRC. As an example, a comprehensive battery of neuroimaging paradigms is being developed that will provide a standard for the assessment of cognitive deficits and residual neural function after serious brain injury. At 3T, they will develop new fMRI methods for real-time communication in the absence of any behavioural response in seriously brain-injured patients.

The team are also developing new techniques for acquiring and combining EEG and fMRI data in real-time, using the ‘dynamic adaptive imaging’ (DAI) technology developed by Rhodri Cusack. At 7T, they are developing new methods for exploring brain damage in these clinical populations using diffusion tensor imaging (DTI) to examine the disruption of white matter fiber tracts in vivo. Somewhat surprisingly, VS and MCS patients spontaneously move quite frequently and dramatically, so the ability to take quick fMRI “snapshots” of the brain activity of these patients is critical and will require the highest sensitivity for single trial fMRI paradigms.

Mild Traumatic Brain Injury

The London Sport Concussion Program (LSCP) brings together clinicians and basic scientists working together to conduct research on the natural history of biomarkers in concussion. The group studies people across the lifespan during which concussion is observed. Specifically, two patient populations are studied. The first is newly concussed individuals who are studied at the time of injury and then again after clinical resolution. The second is varsity team athletes who are at risk for sustaining a concussion.  This study incorporates baseline evaluation and longitudinal follow-up. A major component of both studies is to characterize the structural and metabolic changes in the brain induced by sports-related concussion injury using high resolution anatomical MRI, susceptibility-weighted MRI, MR spectroscopy, resting state fMRI, and diffusion tensor imaging performed using the 3T Siemens MRI scanner at Robarts. In addition, animal studies will be used to validate the results in the athletes and will be used to generate new candidate biomarkers for the human studies.

Preterm Infants and Intraventricular Hemorrhages

This research program is developing new technologies to monitor the brain of preterm infants who are at risk of developing hydrocephalus. Three-dimensional ultrasound and Near-Infrared Spectroscopy have been used at the bedside to monitor changes in ventricular volume as well as their reflection on the cerebral blood flow. The aim of the research is to determine if more accurate metrics can help designing clinical guidelines for the treatment of this population.

Robert Bartha, PhD, MR Physics, Alzheimer Disease
Michael Borrie, MD, Geriatric Medicine
Jorge Burneo, MD, Neurology
Sandrine DeRibaupierre, MD, Neurosurgery
Rhodri Cusack, PhD, Neonatal Neuroimaging
Neil Duggal, MD, Spinal Cord Compression
Roy Eagleson, PhD, Visualization, Psychophysics
Stefan Everling, PhD, Neuroimaging
Aaron Fenster, PhD, 3D Ultrasound
Elizabeth Finger, MD, Cognitive Neurology
Mel Goodale, PhD, Cognitive Psychology
Robert Hammond, MD, Neuropathology
Victor Han, MD, Pediatrics
Mandar Jog, MD, Parkinson’s Disease
Marcelo Kremenchutzky, MD, Multiple Sclerosis
Donald Lee, MD, Neuroradiology
Penny MacDonald, MD, Movement Disorders
Ravi S. Menon, PhD, fMRI, Neuroimaging
Seyed Mirsattari, MD, Epilepsy
Manuel Montero-Odasso, MD, Geriatrics
Adrian Owen, PhD, Brain Injury
Andrew Parrent, MD, Neurosurgery
Stephen Pasternak, MD, PhD, Alzheimer Disease
Terry M. Peters, PhD, Neuroimaging
Jane Rylett, PhD, Neurodegeneration
Kevin Shoemaker, PhD, Neurovascular System
David Steven, MD, Neurosurgery
Keith St. Lawrence, PhD, Near-Infrared Spectroscopy
John Wells, MD, Neurosurgery