A carregar...
Publicação
Optimizing miniature electrodes and current approaches to EEG analysis
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 14.81 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
Many neurodegenerative and neuronal diseases use EEG as a primary diagnosis tool, providing reliable moninvasive real-time evaluation of neuronal processes in the human brain. For epilepsy patients, for example, monitoring the disorder will allow ascertaining if a seldom or rare occurring seizure corroborates a diagnosis or when investigating how many seizures happen during a day and locate them in space to reach a precise diagnostic plan. However a long EEG recording is still unideal for users. Day-long EEG’s involve a bulky, eye-catching medical cap that provokes discomfort after a couple of hours as well as an undesirable appearance to the wearer. The EEG analysis techniques has also been facing novel improvements over the last few decades such as the development sophisticated artifact detection and removal algorithms and improved head models for source estimation. Increasing the wearability of an EEG setup that also elevates the quality of signal using such novel techniques seems to be an important step towards other uses in the research (i.e. Brain Computer Interface applications) and medical communities. The miniaturization of the EEG: both in size and weight but also visual perception, seems to be a logical step to elevate such a system. The idea that one could have smaller, miniaturized electrodes, without any sort of cap; came from the team of Dr. Vadim Nikulin et al[1]. The idealized new electrode system would be a portable, reliable, reusable, invisible, new device able to allow accurate medical diagnosis. Another goal of this dissertation would also be the development of analysis algorithms for EEG applications and studies that accompanies the hardware development and provides the EEG research and medical community with valuable analysis tools. The Electrodes developed are composed of an Ag/AgCI ring with a copper wire and a plasticized cable for easy maneuvering. They demonstrate similar amplitude range signal and SNR as standardly used EEG electrodes. A DFA analysis for High Frequency Oscillations is useful in order to identify common bursts across subjects but also to speculate that the brain could work at a criticality point of the frequency spectrum. SSD is a spatial filtering technique that proved to extract accurately the strongest sources of alpha and theta activity across subjects, thus improving their signal-to-noise ratio. In a visual study of perceptual memory, it provided the conclusion that alpha oscillations not only influence perceptual bias but also the variability in the model and that strong theta oscillations are correlated with less effective visual detection, potentially due to drowsiness. Moreover, it allowed an accurate inverse source modelling analysis with the Brainstorm software, pointing at the occipital cortex as the origin that alpha wave range; while aiming the origin for the theta range frequency for an area concordant with the posterior cingulate cortex, both of these appointed in literature as sources for visual perception [2] [3]. The PCO algorithm could recover both the source pattern of the coupled alpha source as well as the original relation between the phase of the source signal. However, without the SSD preprocessing, PCO is prone to overfitting. PCO hopes to be a valuable tool to increase the sensitivity of phase coupling analyses and to localize the sources of phase coupling. Although this dissertation presents an EEG system composed of miniaturized EEG electrodes as well as several analysis tools, future work needs to be developed in order to confirm the effectiveness of this approach when used on medical patients such as epilepsy patients.
Descrição
Tese de mestrado integrado, Engenharia Biomédica e Biofísica (Engenharia Clínica e Instrumentação Médica) Universidade de Lisboa, Faculdade de Ciências, 2018
Palavras-chave
Electroencephalografia Produção de eléctrodos Criticalidade cerebral Phase coupling optimization Spatial spectral decomposition Detrended fluctuation analysis Teses de mestrado - 2018