Mapping of the sensorimotor cortex: functional MR and magnetic source imaging.

PURPOSE To assess the reliability and comparability of functional MR imaging and magnetic source imaging for mapping the somatosensory cortex. METHODS Parallel studies were performed in eight volunteer subjects in whom both hemispheres were measured with the use of painless tactile stimulation of the tip of each index finger. Magnetic source imaging was performed using a 37-channel biomagnetometer; evoked magnetic fields were analyzed using the single-equivalent dipole representation to ascertain the neuronal source. Functional MR imaging was performed on a 1.5-T MR unit. Blocks of images during periods of rest and activation were acquired using gradient-echo echo-planar imaging. Correlation analysis identified pixels in which signal intensity correlated with the stimulus function. A subsequent requirement for spatial connectivity of activation was imposed to reduce the random occurrence of pixels satisfying the correlation criteria. RESULTS Using temporal and spatial statistical criteria for activation, we found that functional MR imaging showed activation in 1 of 16 hemispheres. In three cases, this was accompanied by activity either frontally or ipsilateral to the stimulus. Magnetic source imaging showed parietal contralateral location in all 16 cases. Where successful localization was achieved with both methods, the separation between sources appeared to be between 1 and 4 cm. Functional MR imaging localizations tended to lie more superficially than the magnetic source imaging localizations. Performance of a simple motor task, rather than use of somatosensory stimulation, resulted in a cortical signal change detectable with a similar functional MR imaging approach in all cases, suggesting the more robust nature of this stimulus. CONCLUSIONS Functional mapping of the somatosensory cortex can be achieved with functional MR imaging or magnetic source imaging. Functional MR imaging yields more spurious locations and fails to show localization more often. If neuronal signal propagation pathways are of interest, the temporal resolution of functional MR imaging alone may be inadequate. A combination of magnetic source imaging and functional MR imaging may allow improved sensitivity, fewer false-positive results, and high spatial and temporal resolution.