Meeting Banner
Abstract #2275

MEG & fMRI: Which One Best Spatially Localizes Brain Activity?

Tynan Stevens1, 2, Steven Beyea, 12, Ryan D'Arcy3, Tim Bardouille2, David Clarke1, 4

1Dalhousie University, Halifax, Nova Scotia, Canada; 2NRC, Halifax, Nova Scotia, Canada; 3Frasier Health Authority, Surrey, British Columbia, Canada; 4Neurosurgery, Capital District Health, Halifax, Nova Scotia, Canada

Functional brain imaging must demonstrate precise functional localization for application to presurgical planning. We investigated two leading candidate technologies - fMRI and MEG - in this context. We implemented a simple grip-force task that has been shown to predominately activate the primary motor region. We evaluated localizing ability using several metrics, including peak activation magnitude (within an anatomically defined ROI), activation extent, laterality, anatomical specificity, and reliability of activated regions. We found that similar peak activation magnitude, lateralization, and reliability could be achieved with either modality. However, fMRI had advantages in terms of specificity to our anatomically defined ROIs.

Abstract PDF

Keywords

abbreviations activated activation active activity advances although amplitude analyzed anatomical appears around assess authority band becomes best blocks brain brainstorm button calculations central chosen closely collected combinations completely conditions constraints context contiguous contrast contrasts corrected cortex created deviation double duration efficacy evaluated event every evoked expect extensive extent fewest filtered filtering flash formed functional functions geometry greater grid hand hands health healthy hemisphere importance increasingly language latency linear localization localize localized localizes localizing magnitude majority mapping maps maximal measured modality model motion motor movement must negative neurosurgery norm nova outside overlap part participant pass patterns peak period polarity popular positive post precision preliminary press presses presumed primary produced produces program regressors reliable represent resolution respectively response rest retest rightward runs sampled segmentation sensitivity separate session significantly simple slices smoothed source spatial spatially specificity statistic stimuli stimulus strength subject summary surrey system table task threshold tools unclear unit unlikely volume volunteers whitened