Robust Characterization of Hippocampus Dynamics from Magnetoencephalography
Zusammenfassung der Projektergebnisse
This project primarily aimed to improve the detectability of hippocampus activity from noninvasive MEG recordings. We were able to develop an easy-to-use-pipeline that constructed boundary element models from structural MR images of participants and generated lead field estimations from them. We combined this with our flagship source-time-frequency reconstructions techniques based on beamforming. We then designed a spatial memory experiment that was run with both healthy participants as well as epilepsy surgery patients who were being implanted with intracranial EEG electrodes. We were furthermore able to successfully acquire 1 dataset of MEG simultaneously with intracranial EEG while the patient performed the spatial memory experiment, an extremely rare but invaluable combination. This data together with the data from the other intracranial EEG recordings served as important validation for the results found with noninvasive MEG. It furthermore enabled a unique kind of analysis, assessing connectivity between deep structures recorded with high-fidelity from intracranial EEG and wholebrain cortical activity reconstructed from MEG. We found that decreases in theta-band activity in the hippocampus and entorhinal cortex appear necessary for spatial memory performance. Our findings indicate that decreased slow-theta activity reflects local and long-range neural mechanisms that encode accurate spatial contexts, and strengthens the view that local suppression of low-frequency activity is essential for more efficient processing of detailed information. We developed techniques for 3D reconstruction of MEG/EEG sensor positions from camera data. This results in improved coregistration to structural MRI data and ultimately improves source reconstruction accuracy and boosts the signal-to-noise-ratio of their time courses. Finally, we developed the foundation for reconstructing hippocampus activity solely from electrocorticographic (ECoG) electrode grids implanted on the cortical surface of epilepsy surgery patients. Many epilepsy surgery clinics around the world implant only ECoG grids for various reasons, although the hippocampus is often a prime suspect in seizure generation; with development, our technique could allow the activity of the hippocampus to be characterized both to inform epilepsy surgery planning and to expand neuroscience research possibilities from the same recordings. We originally expected to design an experiment focusing on increases in hippocampal theta power as had been commonly observed in previous spatial navigation experiments as well as episodic memory experiments. However, in the beginning phases of our experimental design, reductions in this theta power were actually found to be predictive of performance in many kinds of memory experiments; we hypothesized that this would be the case with our spatial memory task, and indeed we confirmed this to be the case.
Projektbezogene Publikationen (Auswahl)
- (2014). Consequences of EEG electrode position error on ultimate beamformer source reconstruction performance. Frontiers in Neuroscience, 8:42. 7pp.
Dalal SS, Rampp S, Willomitzer F, Ettl S
(Siehe online unter https://doi.org/10.3389/fnins.2014.00042) - (2014). Does Illusory Flickering Result from Rhythmic Sampling of Visual Stimuli? J Neurosci, 34:343-345
Crespo-García M, Hartman T
(Siehe online unter https://doi.org/10.1523/JNEUROSCI.4486-13.2014) - (2016). Slow-theta power decreases during item-place encoding predict spatial accuracy of subsequent context recall. NeuroImage, 142:533-543
Crespo-García M, Zeiller M, Leupold C, Kreiselmeyer G, Rampp S, Hamer HM, Dalal SS
(Siehe online unter https://doi.org/10.1016/j.neuroimage.2016.08.021) - (2017). Photogrammetry-based head digitization for rapid and accurate localization of EEG electrodes and MEG fiducial markers using a single digital SLR camera. Frontiers in Neuroscience, 11:264
Clausner T, Dalal SS, Crespo-García M
(Siehe online unter https://doi.org/10.3389/fnins.2017.00264)
