RT Journal Article
SR Electronic
T1 Whole-brain Functional MR Imaging Activation from a Finger-tapping Task Examined with Independent Component Analysis
JF American Journal of Neuroradiology
JO Am. J. Neuroradiol.
FD American Society of Neuroradiology
SP 1629
OP 1635
VO 21
IS 9
A1 Chad H. Moritz
A1 Victor M. Haughton
A1 Dietmar Cordes
A1 Michelle Quigley
A1 M. Elizabeth Meyerand
YR 2000
UL http://www.ajnr.org/content/21/9/1629.abstract
AB BACKGROUND AND PURPOSE: Independent component analysis (ICA), unlike other methods for processing functional MR (fMR) imaging data, requires no a priori assumptions about the hemodynamic response to the task. The purpose of this study was to analyze the temporal characteristics and the spatial mapping of the independent components identified by ICA when the subject performs a finger-tapping task.METHODS: Ten healthy subjects performed variations of the finger-tapping task conventionally used to map the sensorimotor cortex. The scan data were processed with ICA, and the temporal configuration of the components and their spatial localizations were studied. The locations with activation were tabulated and compared with locations known to be involved in the organization of motor functions in the brain.RESULTS: Components were identified that correlated to varying degrees with the conventional boxcar reference function. One or more of these components mapped to the sensorimotor cortex, supplementary motor area (SMA), putamen, and thalamus. By means of ICA components, sensorimotor cortex, supplementary motor area, and superior cerebellar activation were identified bilaterally in 100% of the subjects; thalamus activation was contralateral to the active hand in 80%; and putamen activation was contralateral to the active hand in 60%.CONCLUSION: ICA processing of multislice fMR imaging data acquired during finger tapping identifies the sensorimotor cortex, SMA, cerebellar, putamen, and thalamic activation. ICA appears to be a method that provides information on both the temporal and spatial characteristics of activation. Multiple task-related components can be identified by ICA, and specific activation maps can be derived from each separate component.