Speaker
Description
The non-invasive localisation of active brain regions is not only an important task in basic brain research, but also clinically, for example in the preoperative diagnosis of epilepsy. This is made possible by electroencephalography and magnetoencephalography (EEG/MEG), which measures the electromagnetic field outside the head, from which the underlying current density distribution in the brain can be calculated. This represents a complex inverse problem. Using evoked stimulus responses, source reconstruction methods can be evaluated on real but controlled data sets. The visual evoked response is suitable for estimating active zones in the brain due to the direct topological relation between stimulus and activated cortex volume.
My bachelor's thesis deals with the question of whether different areas in the visual cortex can be reconstructed for different visual angles used for the stimulation. The stimuli are three different visual angles with a flickering dartboard-like pattern in the lower right quadrant. In theory, it is described in such a way that stimuli in the lower right quadrant of the visual field lead to neuronal activity in the upper left of the calcarine sulcus in the primary visual cortex. Thereby, central stimuli (closer to the fixation cross) are located more lateral closer to the skull, whereas peripheral stimuli are located deeper in the brain. A data set of 1000 pattern reversals (on average 2 per second) are recorded per stimulus during simultaneous EEG and MEG measurements. The measurements are conducted on 8 subjects. The data is analysed using the CURRY8 software. The results and further details will be presented in the talk.
