Research Project
9055423
? DESCRIPTION (provided by applicant): There is a great need in clinical neuroscience for improved methods of non-invasively monitoring brain activity. For example, patients with epilepsy often undergo procedures in which electrodes are inserted into their brain to localize the source of seizures. Such invasive procedures carry risk, motivating the need for a non-invasive technique that could detect seizures deep in the brain. One possible technique is an imaging modality known as electrical impedance tomography (EIT), which detects local changes in electrical impedance within the brain that occur during neural activity. Such changes in impedance are detected by injecting small currents through the brain using electrodes that are placed on the scalp. Despite many years of research, EIT has not been adopted for clinical use in neuroimaging because the quality of the images is relatively poor. For example, researchers have found that when a human patient undergoes EIT imaging, it is possible to detect a change in brain activity in response to some event (e.g. a visual stimulus), but the signals are extremely small and it is difficult to localize where in the brain the activity is coming from. Our approach o improving the quality of EIT images is to address what we believe is the core limitation of this technique: the inability to control the path of the injected current through the brain. Specificall, when current is injected into the head via scalp electrodes, the current diffuses widely, and as a result, the measurements reflect any changes in impedance throughout the brain, producing a noisy and imprecise estimate of neural activity. Our hypothesis is that one can improve the quality of EIT images by improving the ability to control the path of the current through the brain We will adopt a technique that is widely used in particle accelerators to control the path of charged particles: we will introduce a magnetic field that will essentially steer the current through the brain in a controlled path. We plan to test the following hypothesis: if current is injected into the brain through scalp electrodes, then the application of a static magnetic field i the direction of the applied current flow will act to confine the current to a small, collimated volume. As a result, fluctuations in voltage that are measured from the scalp electrodes will reflect neural activity from a precise location within the brain. We will test this hypothesis usin both simulations and benchtop testing. The simulations will allow us to understand the relationship between an applied magnetic field and the path of the volumetric current flow through a conductive medium. We will then use these findings to guide the benchtop testing, in which we use experimental models of the head made from a saline tank and a plaster model of the skull. Our proposed technique of steering currents through the brain with magnetic fields has the potential to significantly improve the quality of images obtained by EIT, yielding a potentiall powerful technique for non-invasive monitoring of neural activity.
