Desenvolvimento de uma plataforma para o processamento estatístico de sinal eletroencefalográfico (EEG)

Abstract

The aim of this work is the development of a software for the statistical analysis of electroencephalography (EEG) signals, focusing on Descriptive Statistical Analysis and also on Statistical Tests, such as Hypothesis Testing and Normality Tests, thus, all EEG registers (containing 20 channels and N periods) used in the software were submitted to these statistical analyzes that previously had information on three different quantifiers, which are: Power Contribution Percentage (PCP) (informs the amount of energy contained in a band of frequency); Median Frequency (FM) (value of the average frequency in the frequency range) and finally Coherence (degree of similarity between the right and left hemispheres). As a result of the study, a platform was obtained with seven interfaces in total, distributed in: Main Interface, Descriptive Statistical Analysis Interface, Visualization Interface and Test Interfaces (Hypotheses and Normality). The software presented here was tested only with actual EEG records of comatose patients and normal volunteers, and all data acquired in the statistical analysis were compared with renowned statistical software on the market. The validation of the software was performed by comparing the data obtained in this work with the results obtained in two other software, comparing values from descriptive statistical processing, normality tests and hypotheses, and all comparisons were satisfactory and the same as the others comparative software. Therefore, the software of this work proves to be of great value, given that it is exclusively focused on statistical analysis of EEG records, obtained a better processing time when compared to other software used, is easy to use and very practical, besides to avoid the use of several platforms to perform the statistical analysis, since in this software it is possible to perform a satisfactory and complete statistical analysis.