Research


Magnetic Resonance Imaging Repository

Over 30 years, MS/MRI has become MRI analysis centre and compiled scans from local, national, and international institutions and practices to create a comprehensive source of conventional MRI images as well as Advanced MRI images of people with MS. Conventional MRI is a widely used technique for the diagnosis and monitoring of disease progression in people living with MS. MRI sequences such as FLAIR, 3D T1, PD, and T2 are well established and routinely used in combination with image analysis tools in research to help identify brain and spinal cord lesions, atrophy, cortical thickness, brain parenchymal fraction, burden of disease, and corpus callosum area.

Currently we are developing an Imaging Repository that will allow researchers to deposit, data-mine, share, and interact with large imaging datasets of the brain and spinal cord from people with MS.The Imaging Repository will have highly scalable capacity to store and share raw and processed multi-modal images in commonly used imaging formats (e.g., DICOM, NIfTI).

Current registry projects:

  • Canadian Prospective Cohort Study to Understand Progression in Multiple Sclerosis (CanProCo)
  • The North American Registry for Care and Research in Multiple Sclerosis (NARCRMS)
  • North American Imaging in MS Cooperative – Imaging Repository (NAIMS-IR)
Advanced Magnetic Resonance Imaging

Advanced MRI, unlike conventional MRI, allows for quantitative imaging, which can accurately and objectively measure intrinsic properties. Myelin water imaging (MWI) is an advanced MRI technique that resolves the portion of MR signal associated with water trapped between the myelin bilayers. MWI has been validated with both human histology and animal models as a specific measure of myelin. Magnetization Transfer Imaging (MTI/MTR) also evaluates myelin integrity. Diffusion tensor imaging (DTI) and Diffused Based Spectrum Imaging measure the diffusion of water in brain tissue. Susceptibility-weighted imaging (SWI) is a technique that is sensitive to myelin and iron, and is used to identify “rim” lesions. MR spectroscopy measures brain metbaolites that can be lowered with damage (NAA) or increased with inflammation and scarring (ml, Lactate)

Current Projects:

  • iCAMMS: neuroprotective effects of alemtuzumab on advanced MRI outcomes assessed over 2 years (completed)
  • OPERA: neuroprotective effects of ocrelizumab and interferon on advanced MRI outcomes assessed over 2 and 5 years (ongoing)
  • OBOE: longitudinal MRS study in ocrelizumab treated patients compared to serum biomarkers
  • VELOCE: longitudinal DBSI study in ocrelizumab treated patients
  • LEMCOG: relationship between myelin imaging and longterm cognitive outcome in alemtuzumab treated MS patients
  • LOBSTr: longitudinal natural history study of brain and spinal cord myelin imaging in MS (MS society and CIHR funded)
Machine Learning

Machine learning (ML) uses statistical models and algorithms that allow computers to perform specific tasks without explicit instructions. The programmed algorithms enable software applications to predict outcomes with high accuracy. Deep learning is a machine learning approach that uses layered hierarchical, graphical networks to extract features from data at progressively higher levels of abstraction. Unsupervised deep learning can be particularly useful in neuroimaging, a domain in which the number of labeled training images is typically limited.

Sample average images of quantitative brain images in a group of healthy adults (average age 35 years and a range of 20-53 years):

O’Muircheartaigh J, Vavasour I, Ljungberg E, Li DKB, Rauscher A, Levesque V, Garren H, Clayton D, Tam R, Traboulsee A, Kolind S. 2019. Quantitative neuroimaging measures of myelin in the healthy brain and in multiple sclerosis. Hum Brain Mapp. 2019 Jan 15.