Next generation neuroimaging using optically pumped magnetometers

点击翻译按钮获取中文摘要

Much of what we understand about the human brain comes from functional neuroimaging - a collection of extremely powerful non-invasive techniques which can measure brain activity in real time, such as functional magnetic resonance imaging (fMRI) or magnetoencephalography (MEG). However, perhaps the biggest limitation of these techniques is that they are heavily reliant on subjects lying very still within large and cumbersome scanners. For example, MEG measures magnetic fields outside the head generated by current flow in the brain. In current devices, the requirement for cryogenically cooled magnetic field sensors means that those sensors are fixed in position, and any head movement during scanning degrades the quality of the data. The requirement that subjects remain still limits both the subject cohorts that can be scanned (e.g. it is hard to scan children) and the types of experimental paradigm that can be accessed (e.g. it is challenging to measure brain activity during natural tasks such as learning to play a musical instrument, or whilst a subject is immersed in virtual reality). These limitations are however being lifted by the introduction of quantum technology to functional imaging. New quantum enabled 'wearable' MEG systems are now under development in Nottingham and elsewhere. These systems are built around optically pumped magnetometers - small and lightweight field sensors that can be mounted directly on the head. Flexibility in sensor placement means that systems can be adapted to any head shape or size. Further, if background fields are appropriately nulled, subjects can move freely during data collection. Though still a nascent technology, these new wearable scanners are making significant headway in enabling a new generation of neuroscientific experimentation. However, little is yet known about their fundamental capabilities. In this Ph.D., we will work on the development, characterisation, and application of Nottingham's wearable MEG device, with a view to turning it into a commercial device. At a fundamental level we will work on how to deploy OPM sensors in an optimised array around the scalp to maximise both coverage and spatial resolution. Using techniques including blind source separation we will characterise spatial resolution and its dependence on sensor separation, sensor cross talk (which we will mathematically model), and signal to noise ratio (the latter being improved by reference array measurements). We will work on characterising the bandwidth of the OPM based scanner: Bandwidth is particularly important for clinical studies; here we will look to very high frequency measurements, attempting to measure signals outside what is conventionally thought to be the higher end of the bandwidth of an OPM. Finally, we will develop new experimental paradigms. Specifically, using a three-projector set up and polarised light, we will build a virtual reality CAVE inside Nottingham's newly installed magnetically shielded room. We will then use this to assess how the human brain responds to making decisions in stressful situations.