ABSTRACT
Resumen: El Potencial de disparidad es una respuesta cortical elicitada por la detección automática de estímulos con distintas características, permitiendo la exploración de procesos neuropsicológico. Sin embargo el análisis de esta señal se puede dificultar por una baja relación señal a ruido debida a los artefactos presentes en la adquisición de la misma. Diversas publicaciones proponen el uso de implementaciones de la técnica de Separación Ciega de fuentes, como el Análisis por Componentes Independientes (ACI), para preprocesar las señales y eliminar estos artefactos. Sin em bargo, no se ha estudiado cuál de los algoritmos ACI que se encuentran en la literatura será el óptimo para mejorar la calidad del MMN, por lo que en este estudio se propuso determinar si existen diferencias significativas en las respuestas obtenidas al utilizar los algoritmos de FastICA, Infomax y SOBI para eliminar los artefactos típicamente presentes en este tipo de señales. Adicionalmente se dan algunas características de estos artefactos a manera de sistematizar la identificación y eliminaciones de los mismos, además de comparar las respuestas obtenidas con y sin preprocesamiento, así como la distribución topográfica de este potencial antes y después de la eliminación de artefactos. Mediante el algoritmo Infomax se identifican mejor los Componentes Independientes asociados con artefactos, resultando en un MMN de mayor amplitud y distribución topográfica fronto-central con predominancia izquierda.
Abstract: Mismatch Negativity is a cortical response elicited by the automatic detection of stimuli which have different characteristics, allowing exploration of neuropsychological processes. However, the analysis of this signal can be di fficult by a low SNR due to artifacts present when the signal is recorded. Different publications propose to use the approach given by the Blind Source Separation Technique by means of the Independent Component Analysis (ICA) to preprocess and eliminate these artifacts. Nevertheless, it has not been studied which of the ICA algorithms found in the literature will be optimal for improving the quality of MMN. Therefore the aim of this study is to determine whether there are significant differences in the responses obtained by using FastICA, Infomax and SOBI to remove artifacts typically present in such signals. In addition, some features of the Independent Components related to artifacts are given in order to systematize the identification and elimination of those. In addition, MMN responses obtained with and without data preprocessing, as well as topographic maps before and after the elimination of artifacts were compared. Thus, Infomax is the best ICA algorithm to calculate Independent Components associated with artifacts, resulting in high amplitude MMN and a topographic map with a clear fronto-central distribution with left-hemisphere predominance.
ABSTRACT
Although neuroimaging techniques and other diagnostic procedures has been developed, electroencephalography(EEG) is still very important for the evaluation of various brain diseases and functional studies of human brain. EEG is formed mainly by spatial and temporal summations of postsynaptic potentials generated from a large population of pyramidal cells that can be considered as a collection of oscillating dipoles. EEG shows continuous rhythmic oscillation depending on sleep-waking state. Alpha rhythms are generated in cortical areas acting as epicenters with local spread, although the precise cellular mechanism is still unknown. It's been known that neurons in the nucleus reticular thalami are the pacemakers of sleep spindle. Alterations in the circuit of the reticular nuclei-thalamocortical relay neuron-cortical neuron are responsible for generalized spike and wave complexes. At the intracellular level, large paroxysmal depolarizing shifts produce focal epileptic spikes. Slow waves of EEG appear to be related to thalamocortical and/or corticothalamic deafferentation. The interpretation of routine EEG requires a well training from a qualified EEG teacher and reading adequate amount of EEG under supervision. Frequent misinterpretations of routine EEG have been observed in both local clinics and general hospitals. The most common findings of normal routine EEG misinterpreted as abnormal are normal variants and artifacts of various sources. There are considerable variations of normal EEG rhythms and pseudoepileptiform discharges. Eyeball movements produce prominent or subtle EEG changes over the frontal regions that are sometimes hard to be differentiated from abnormal slow waves over that region. Systematic approach was described for a good interpretation of routine EEG.