Reliable recording and analysis of MEG-based corticokinematic coherence in the presence of strong magnetic artifacts. Issue 2 (February 2016)

• Record Type: Journal Article
• Authors: Bourguignon, Mathieu; Whitmarsh, Stephen; Piitulainen, Harri; Hari, Riitta; Jousmäki, Veikko; Lundqvist, Daniel
• Is Part Of: Clinical neurophysiology. Volume 127:Issue 2(2016:Feb.)
• Page Start: 1460
• Page End: 1469
• Publication Date: 2016-02
• Subjects: Magnetoencephalography -- Artifacts -- Signal-space separation -- Corticokinematic coherence -- Human brain -- Proprioception -- Functional mapping; Neurophysiology -- Periodicals; Electroencephalography -- Periodicals; Electromyography -- Periodicals; Neurology -- Periodicals; 612.8
• DOI: 10.1016/j.clinph.2015.07.030
• Languages: English
• ISSNs: 1388-2457

Abstract

Highlights: MEG-based corticokinematic coherence (CKC) reliably locates the primary sensorimotor (SM1) cortex. In subjects wearing magnetized material, temporal signal-space separation effectively cleans the MEG data. In the presence of such artifacts, CKC still locates the SM1 cortex with ∼5 mm accuracy and allows reliable studies of proprioception. Abstract: Objective: Corticokinematic coherence (CKC) is the coupling between magnetoencephalographic (MEG) signals and limb kinematics during fast movements. Our objective was to assess the robustness of CKC-based identification of the primary sensorimotor (SM1) cortex of subjects producing strong magnetic artifacts when the MEG signals were cleaned with temporal signal space separation (tSSS). Methods: We recorded MEG during active and passive forefinger movements and during median-nerve stimulation in the following conditions: (1) artifact-free, (2) a magnetic wire attached to the scalp at C3 location, and (3) a magnetic wire attached behind the lower central incisors. Data were pre-processed with tSSS and analyzed using standard CKC methods, somatosensory evoked fields (SEFs), and dipole modeling. Result: Artifacts were effectively suppressed by tSSS, enabling successful identification of the SM1 cortex in all subjects based on CKC and SEFs. The sources were in artifact conditions ∼5 mm away from the sources identified in artifact-free conditions. Conclusion: tSSS suppressed artifacts strongly enough to enable reliable identification of the SM1 cortex on the basis of CKC mapping, with localization accuracy comparable to SEF-based mapping. Significance: The results suggest that CKC can be used for SM1 cortex identification and for studies of proprioception even in patients implanted with magnetic material.