Principles and Applications of Magnetoencephaloraphy (MEG) for Non-Invasive Brain Imaging

Advances in brain imaging technology have provided new insight about how our brain operates and how we can improve the diagnosis and treatment of neurological diseases. Studying normal processes or pathological conditions like epilepsy requires the detection of neuronal signals at a very high temporal resolution. Invasive neurophysiological techniques can directly detect the associated neural currents, but non-invasive methods would presumably carry less risk to patients and be more applicable to healthy subjects. The best temporal resolution in non-invasive recordings is achieved by directly measuring the electromagnetic fields produced by these neural currents. Electroencephalography (EEG) measures the voltage distribution on the head surface and magnetoencephalography (MEG) detects the magnetic field outside of the head. In this talk, I will review the basic principles of some commonly used non-invasive neuroimaging techniques and show that MEG has the best combination of both spatial and temporal resolution. MEG relies on basic principles of quantum mechanics and theorems derived from Maxwell's equations to overcome the challenge of detecting extremely weak biomagnetic signals and constrain solutions posed by the inverse problem. Signal processing methods for interference suppression will be explained, which are crucial especially in clinical measurements where tremors, involuntary movements, stimulators, and pacemakers may generate large artifacts that compromise the data quality and demonstrate how these methods can be extended to compenstate for movement-related field distortions in experiments with awake infants and children. Furthermore, I will focus on some of the recent mathematical developments and future prospects that may significantly widen the applicability and popularity of MEG in both basic research and clinical settings. Examples of applications will be shown that emphasize the unique benefits of MEG.