ABSTRACT
OBJECTIVE: To present the technical design and demonstrate the feasibility of a multi-channel on-scalp magnetoencephalography (MEG) system based on high critical temperature (high-[Formula: see text]) superconducting quantum interference devices (SQUIDs). METHODS: We built a liquid nitrogen-cooled cryostat that houses seven YBCO SQUID magnetometers arranged in a dense, head-aligned array with minimal distance to the room-temperature environment for all sensors. We characterize the performance of this 7-channel system in terms of on-scalp MEG utilization and present recordings of spontaneous and evoked brain activity. RESULTS: The center-to-center spacing between adjacent SQUIDs is 12.0 and 13.4 mm and all SQUIDs are in the range of 1-3 mm of the head surface. The cryostat reaches a base temperature of â¼ 70 K and stays cold for 16 h with a single 0.9 L filling. The white noise levels of the magnetometers is 50-130 fT/Hz1/2 at 10 Hz and they show low sensor-to-sensor feedback flux crosstalk ( 0.6%). We demonstrate evoked fields from auditory stimuli and single-shot sensitivity to alpha modulation from the visual cortex. CONCLUSION: All seven channels in the system sensitively sample neuromagnetic fields with mm-scale scalp standoff distances. The hold time of the cryostat furthermore is sufficient for a day of recordings. As such, our multi-channel high-[Formula: see text] SQUID-based system meets the demands of on-scalp MEG. SIGNIFICANCE: The system presented here marks the first high-[Formula: see text] SQUID-based on-scalp MEG system with more than two channels. It enables us to further explore the benefits of on-scalp MEG in future recordings.
Subject(s)
Magnetoencephalography , Scalp , Animals , Brain , DecapodiformesABSTRACT
OBJECTIVE: We present a benchmarking protocol for quantitatively comparing emerging on-scalp magnetoencephalography (MEG) sensor technologies to their counterparts in state-of-the-art MEG systems. METHODS: As a means of validation, we compare a high-critical-temperature superconducting quantum interference device (high Tc SQUID) with the low- Tc SQUIDs of an Elekta Neuromag TRIUX system in MEG recordings of auditory and somatosensory evoked fields (SEFs) on one human subject. RESULTS: We measure the expected signal gain for the auditory-evoked fields (deeper sources) and notice some unfamiliar features in the on-scalp sensor-based recordings of SEFs (shallower sources). CONCLUSION: The experimental results serve as a proof of principle for the benchmarking protocol. This approach is straightforward, general to various on-scalp MEG sensors, and convenient to use on human subjects. The unexpected features in the SEFs suggest on-scalp MEG sensors may reveal information about neuromagnetic sources that is otherwise difficult to extract from state-of-the-art MEG recordings. SIGNIFICANCE: As the first systematically established on-scalp MEG benchmarking protocol, magnetic sensor developers can employ this method to prove the utility of their technology in MEG recordings. Further exploration of the SEFs with on-scalp MEG sensors may reveal unique information about their sources.