A new hardware controller will sample magnetic fields at high frequency and calculate cancelling currents fast enough to suppress sudden magnetic disturbances, making quantum brain scanners practical outside specialised labs. Current magnetically shielded rooms for OPM-MEG (optically pumped magnetometer magnetoencephalography) require multiple metal layers and active cancellation, making them too heavy and expensive for routine hospital use. The researcher is building a controller that works with a single mu-metal layer, compensating for the lost passive shielding with faster, smarter active cancellation—particularly for rapidly oscillating fields that existing systems struggle to handle. If the controller succeeds, it could slash the cost and weight of shielded rooms, allowing OPM-MEG to move from a handful of specialist centres into general clinical settings for epilepsy, dementia, and brain injury diagnosis. The low-latency system also opens the door to real-time brain-state experiments, where stimuli are presented in response to a person’s ongoing neural activity. Separately, the researcher plans to apply machine learning to the group’s newly large OPM-MEG datasets, potentially streamlining data processing and revealing patterns invisible to conventional analysis.
View original technical description
My initial research will focus on the implementation of low-latency data processing and hardware control, with the aim of increasing the applicability of OPM-MEG in clinical settings. One area in which this work is pertinent is as part of a broader effort to improve the practical feasibility of installing magnetically shielded rooms: presently, the rooms consist of multiple layers of metal and an active magnetic field cancelling system; because of this, the rooms cost and weigh too much to be used in a broader clinical context. As such, work is being carried out to develop rooms which make use of a single layer of mu-metal, thereby requiring a far more effective active cancelling system, particularly at higher frequencies where the fields oscillate quickly. To address this issue, I am working on producing a hardware controller capable of sampling the signal at a high frequency while calculating the current to be applied to a set of coils to cancel out the magnetic field in a time frame small enough to adequately respond to sudden perturbations in the magnetic field. The hope is that once this task is completed, it will provide a gateway into other applications of low-latency systems, for example, it may be possible to produce a system capable of interpreting collected brain data quickly enough to allow for the presentation of stimuli in response to specific brain states in an experimental setting. It is also hoped that my background in computer science could be applied to the processing of data collected by OPM-MEG. Hitherto, there was a dearth of such data, but the research group is now in possession of large data sets collected across many studies. It would therefore be apposite to consider the application of certain computational techniques - such as machine learning - on OPM-MEG data, both to streamline the act of processing the data, and to potentially glean new insights.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know