Conformal pediatric whole-head MEG system with optically-pumped magnetometers



 DESCRIPTION (provided by applicant): There are major gaps in our understanding of the development of normal and abnormal brain function in the critical years from birth to preschool. These gaps are due in large part to limited methods for non-invasively examining the evolution of dynamic spatiotemporal interactions between brain regions during information pro- cessing, where small head size rules out magnetoencephalography (MEG), leaving only electroencephalography (EEG) with relatively poor spatial resolution. We propose to develop a MEG system that could be able to detect somatosensory evoked magnetic field (SEF) in children produced by tactile stimulation on single trials without signal averaging. We will evaluate the ability to better separate discrete cortical sources with the proposed system as compared to electroencephalography (EEG) and the best pediatric MEG system available (babyMEG). The proposed system is based on uncooled microfabricated optically-pumped magnetometers (µOPM) instead of the Superconducting QUantum Interference Devices (SQUIDs) used in conventional MEG systems. The new type of sensors could enable: 1) a MEG that can fit the head of any size and shape, 2) minimal gap (< 4mm) between each µOPM probe and the scalp for increased signal strength, and 3) a small footprint allowing very close sensor spacing (~15 mm) for improved spatial resolution, and 4) manufacturing on large wafers in parallel, which can reduce operating and system cost. We envision a system in the future, which is much more similar to and EEG system with respect to cost, versatility, and ease of use. We have 3 specific aims: Aim 1A: Develop a highly sensitive 1-channel µOPM MEG system with a magnetometer having <7 fT/vHz noise level and 200 Hz bandwidth. Aim 1B: Implement a µOPM gradiometer with a baseline of 2 cm, a noise level of <10 fT/vHz, and a common-mode rejection ratio (CMRR) of 150. Aim 2A: Develop a 64-channel gradiometer MEG system. We will refine our method for large-scale fabrication of the µOPM probes, emphasizing simplification for cost reduction, reproducibility, and quality control by developing a parallel fabrication process instea of a serial one. Aim 2B: Assemble and test the 64-channel µOPM MEG system on a phantom and healthy adult volunteers in a standard 2-layer MSR at the National Institute of Standards and Technology (NIST). Aim 3A: Evaluate the 64-channel system in healthy children (0-36 months) at the MEG facility of Boston Children's Hospital (BCH) with a focus on detecting SEFs in single trials and determining localization errors. Aim 3B: Evaluate the ability of the system to localize interictal generators in epilepsy patients (1-3 years) and separate multiple generators. By directly comparing data from the µOPM MEG and a 375-channel pediatric MEG system ("babyMEG") based on SQUIDs from healthy children and epilepsy patients we expect to demonstrate not only the feasibility of µOPMs for MEG, but also better MEG images paired with practical advantages.