Advanced functional MRI

Abstract Functional Magnetic Resonance Imaging (fMRI) has been widely used to study the responses of the somatosensory cortex, motor cortex and associated neuronal activity in the human cerebral cortex. fMRI is a non-invasive and indirect method for mapping brain activity through measurement of the hemodynamic responses associated with electrical neuronal activity and the neural activity leads directly to changes in blood flow, blood volume and the cerebral metabolic rate of oxygen consumption. Non-invasive neuroimaging technologies such as functional MRI have both advantages, such as good spatial resolution, and disadvantages, such as poor temporal resolution. Some of the disadvantages have been alleviated by incorporating other techniques such as optical spectroscopy or electroencephalography (EEG) which are also non-invasive. All these techniques are sensitive to the vascular response of neuronal activity but in addition we are now investigating the existence of a weak direct electromagnetic effect with advanced fMRI. This neuronal current effect which gives rise to main magnetic field modulation should provide additional information for studying nerve characteristics. In this thesis, methods for fMRI mapping of responses from phantoms, the median nerve, the visual system, the motor sensory cortex and the thalamus are optimised and subsequently quantified. The experimental results strongly support the main hypothesis of the thesis and suggest that the generated magnetic field due to ionic current can be detected by present generation MRI using specific experimental designs and stimulation paradigms. Overall our results show that ionic currents in subjects can generate percentage signal changes in MRI up to 0.1± 0.01% corresponding to mean magnetic axonal fields of 0.7± 0.1nT with a Signal to Noise Ratio (SNR) of 3:1. The responses of the median nerve, motor sensory cortex and thalamus were detected using transcutaneous electrical nerve stimulation (TENS) and the visual cortex using strobe light stimulation in the range of frequencies 2.1 Hz to 4.1 Hz. All these measurements were acquired at 1.5T. Fast fMRI experiments using TENS and finger tapping were also acquired simultaneously. In addition, real and imaginary finger tapping experiments were performed in the motor sensory cortex at 3T. Our results imply that axonal fields that are generated due to action potentials can generate effects on MRI sensitive enough to directly detect neuronal activity using advanced fMRI, although sensitivity is still not fully adequate for clinical use.