During 2003-2004, the work of AMRI has focused on 3 initiatives: 1) the improvement of of contrast-to-noise ratio (CNR) and resolution in MRI of brain; 2) the development of stimulation protocols that allow the separation of multiple brain processes as detected by BOLD fMRI; 3) the investigation of brain activity patterns during rest and sleep. CNR was improved by development of a new contrast preparation technique that consists of multiple MRI inversion pulses. The technique was optimized with simulated annealing methodology. Experiments performed at 3.0 T have demonstrated excellent separation of grey matter, white matter and CSF. This was achieved at an isotropic resolution of 1x1x1 mm, and in a scan time of 10-minutes. BOLD fMRI allows the measurement and mapping of activity induced by a single task. In order to separate brain activity signals that originate from different brain processes associated with different components of the task, a stimulation protocol was developed that was based on the binary m-sequence. This protocol was tested with a Sternberg task to investigate the relative contribution of controlled and automatic processing for letter recognition. The results have shown that it is indeed feasible to simultaneously measure contribution of controlled and automatic processing in a single experiment. During rest, substantial fluctuations in brain metabolism occur. These are reflected in BOLD fMRI signals, that often show regionally correlated signal fluctuations. These signals are attributed to ongoing activity, possible caused by active though processes, or fluctuations in vigilance. Using highly sensitive MRI detectors, we have studied these fluctuations during extended resting periods, as well as early sleep. It was found that these fluctuations involve all of cortex as well as deep nuclei. Furthermore, their amplitude is increased during early sleep. It is hypothesized that these patterns represent synaptic consolidation processes, and that they have potential value in the study of neuronal viability. We plan to further study these patterns and correlated them with electrical measures of neuronal activity.
Showing the most recent 10 out of 34 publications
