Functional Magnetic Resonance Imaging Research

Functional magnetic resonance imaging, or fMRI, refers to a collection of MRI techniques that have been developed to enable non-invasive visualization of how neurons within the brain become active in response to perceiving a sensory stimulus or specific behaviours such as thinking, remembering, and acting. The technique has become one of the essential tools in basic and applied neuroscience, with more than 1000 papers published each year involving fMRI. Medical applications for fMRI are also emerging, such as a role in neurosurgical planning to enable neurosurgeons to resect pathological brain tissue while sparing as much normal surrounding tissue as possible. Situated at the Rotman Research Institute, a major University of Toronto neuroscience facility, research in my laboratory primarily involves improving biophysical understanding of fMRI signals, as well as technical development to expand fMRI capabilities and improve the quality of fMRI data. The research can be grouped under four broad themes:

1. Comparing fMRI with other functional neuroimaging techniques. The fMRI signal indirectly measures neuronal activity through "neurovascular coupling": the metabolic demands of activated neurons cause local, transient increases in blood oxygenation, flow, and volume that can be measured by MRI. The approach taken in the lab is to measure additional neuronal activity and hemodynamic variables to determine how the spatiotemporal variations in these parameters modulate fMRI data. For example, experiments currently underway involve comparing fMRI data with magnetoencephalography (MEG) data that probe the evoked magnetic fields that accompany neuronal activity at millimeter spatial resolution and millisecond temporal resolution. Ultimately, this work may lead to improved visualization of neuronal activity through mathematical combination of multimodal functional neuroimaging data.

2. Motion artifacts.
Unfortunately, the fMRI signal is extremely susceptible to noise generated by small (millimetre) amounts of head motion, and a number of strategies are under investigation in the laboratory to address the problem. Two examples include: the development and evaluation of an "fMRI simulator system" to exclude patients with excessive head motion and to improve training of participants to remain still; and development of "prospective motion correction" a technique in which head motion is measured enabling fMRI data to track with moving anatomy in real time. The latter technique may help to make fMRI more robust in patient opulations.

3. Virtual Reality.
Virtual reality (VR) is increasingly used to provide better assessment of patient's behavioural characteristics under real-world conditions, and as a cognitive or physical therapy tool to promote recovery from neurodisability such as stroke, or traumatic brain injury. Functional MRI research is being directed towards understanding whether brain activity in VR is similar to that in the real world, and understanding how VR therapy can change brain function. Recent work in this area has focused on the technical development of a prototype MRIcompatible data glove with force feedback capability that enables sensation of touching virtual objects, and comparing the brain activity associated with touching a virtual object with that obtained when touching objects in the real world. This work may ultimately have application to VR assessment and therapy of stroke patients with impaired skin sensation.

4. fMRI of Stroke Recovery.
Stroke is the main neurologic deficit in the elderly, placing a burden of billions of dollars on Canadian society. Although there is much research in prevention and immediate treatment of acute stroke, rehabilitation research receives much less attention. The brain has an innate capacity to reorganize after stroke, so that some stroke victims recover well whereas others do not. The use of fMRI to quantify changes in brain activation with time is critical to understanding stroke recovery mechanisms. Several studies are ongoing in the lab, in collaboration with scientists in the Heart and Stroke Foundation Centre for Stroke Recovery, located at Sunnybrook Health Sciences Centre, the Rotman Research Institute, and the University of Ottawa. Part of the work involves expanding the scope of fMRI applications involving stroke patients, such as development and optimization of experimental protocols to investigate brain activity associated with sensorimotor tasks involving the lower limb, and development of a stylus for using fMRI studies of drawing and impairment of visually guided actions.

For more information, go to the Rotman Research Institute website (http://www.rotman-baycrest.on.ca/index.php?section=208).

Graduate Students:

• Mark Chiew - real-time fMRI technology
• Mahta Karimpoor
• Daivd Rotenberg
• Norman Weir

Selected References:

• Di Diodato LM, Mraz R, Baker SN, Graham SJ. A haptic force feedback device for virtual reality-fMRI experiments. IEEE Trans Neural Syst Rehabil Eng. 2007 Dec;15(4):570-6.

• Sörös P, Marmurek J, Tam F, Baker N, Staines WR, Graham SJ. Functional MRI of working memory and selective attention in vibrotactile frequency discrimination. BMC Neurosci. 2007 Jul 4;8:48.

• Ferber S, Mraz R, Baker N, Graham SJ. Shared and differential neural substrates of copying versus drawing: a functional magnetic resonance imaging study. Neuroreport. 2007 Jul 16;18(11):1089-93.

• Macintosh BJ, Mraz R, McIlroy WE, Graham SJ. Brain activity during a motor learning task: an fMRI and skin conductance study. Hum Brain Mapp. 2007 Dec;28(12):1359-67.

• Gladstone DJ, Danells CJ, Armesto A, McIlroy WE, Staines WR, Graham SJ, Herrmann N, Szalai JP, Black SE; Subacute Therapy with Amphetamine and Rehabilitation for Stroke Study Investigators. Physiotherapy coupled with dextroamphetamine for rehabilitation after hemiparetic stroke: a randomized, double-blind, placebo-controlled trial. Stroke. 2006 Jan;37(1):179-85. Epub 2005 Dec 1.

• Nangini C, Tam F, Graham SJ. A novel method for integrating MEG and BOLD fMRI signals with the linear convolution model in human primary somatosensory cortex. Hum Brain Mapp. 2008 Jan;29(1):97-106.

• Tremblay M, Tam F, Graham SJ. Retrospective coregistration of functional magnetic resonance imaging data using external monitoring. Magn Reson Med. 2005 Jan;53(1):141-9.