Quantum Physics and Computation for Neuroscience: Magnetoencephalography

A key neuroscience technology called Magneto-Encephalography (MEG) provides the highest spatial and temporal resolution of non-invasively recording brain activity. MEG uses Super-conducting Quantum Interference Devices (SQUIDs) as magnetic field sensors to capture emitted neuronal magnetic fields from the brain. SQUID brain recordings have temporal and spatial resolution on the order of one millisecond and one millimeter. The vast data sets from MEG are both in a form and are of such a nature that they could directly benefit from Quantum computing (QC). Further, the physical form of the MEG sensor and quantum nature of their outputs suggests that an appropriate fusion of the sensor to a quantum computing platform will yield efficiencies in both signal and data “impedance” matching, with a positive cost improvement of both completion time and computational accuracy. Of note, the SQUID Josephson Junctions are super-cooled in dewars that are the same as those for current superconducting Qubit computing chips. We will discuss our ongoing end-to-end cost analysis of embedding MEG SQUIDs with QUBIT processors in the same container. This analysis will benefit from and complement a parallel effort of design of QC algorithms for MEG using both noisy intermediate scale quantum (NISQ) and future QCs. Our initiative aims to explore the potential integration of MEG and QC to examine brains in operation at unprecedented speeds and resolution, and to enable computations regarding brain operations never before realized.