Simultaneously Acquired Magnetoencephalography and Diffuse Optical Tomography Data Reveals Correlated Somatosensory Activity

Abstract

Simultaneous measurement of electrophysiological and hemodynamic brain signals imposes special requirements on the instrumentation. Here, we developed a high-density fiberoptic probe for concurrent diffuse optical tomography (DOT) and magnetoencephalography (MEG) recordings. Transparent two-component silicone was mixed with carbon black dye to achieve a black, flexible, non-magnetic support for the dense optode arrangement and low (5 mm) probe thickness. The probe was used to record somatosensory responses to electrical right median nerve stimulation at 0.5, 1, 2, and 4 Hz in 18 adult human subjects. Brain activity was simultaneously measured with a commercial whole-head MEG system and with the DOT optode arrangement covering approximately 40 cm² over the parietal region in the contralateral left hemisphere. Two correlation-based clustering methods were developed to find regions where the reconstructed time course of total hemoglobin concentration (HbT) changes correlated with the predicted hemodynamic activity based on time-course characteristics of the MEG sources and the canonical hemodynamic response model. Two statistically significant clusters were found based on the correlation between HbT around the postcentral gyrus and MEG primary somatosensory cortical activity at ~35 ms (P35m response). In addition, correlation between HbT and secondary somatosensory cortical activity suggested a statistically significant cluster in the postcentral gyrus and parietal operculum. These results illustrate an improvement in localization over previous DOT studies using sparse optode arrangements, and demonstrate the feasibility of the system for simultaneous HD-DOT-MEG experiments. Furthermore, the techniques described here pave the way for understanding the coupling between hemodynamic and electrophysiological responses. Further research is needed to reveal the neuronal circuits giving rise to the correlating MEG and DOT response features. Significant improvements in the technology are still expected via optimization of the detected light power in the instrumentation.