Analysis of pain-related somatosensory evoked magnetic fields using the MUSIC (multiple signal classification) algorithm for magnetoencephalography

To compute an electrical current source from electrophysiological date, the single equivalent current dipole has been widely used. However, the method is not suitable for analyzing multiple dipoles. We used MUltiple SIgnal Classification (MUSIC) algorithm for analyzing multiple dipole sources in pain-related somatosensory evoked magnetic fields (SEFs) that had been obtained by 148-channel whole head type magnetoencephalography. Both middle latency (contralateral SⅠ) and long latency (bilateral SⅡ and cingulate gyri) components were clearly identified in the maps superimpksed on MR images. Not only the MUSIC results were in very good agreement with the single dipole estimation results, but also gave the temporal profiles of each current dipole source in stacks of consecutive maps. Thse results suggested that MUSIC algorithm is an effective tool for the analysis of multiple dipole sources of pain-related SEFs.

原著