Deficits in social interaction and communication characterize Autism Spectrum Disorder (ASD). Although widely recognized by its symptoms, diagnosing ASD remains challenging due to its wide range of clinical presentations. Methods: In this study, we propose a method to assist in the early diagnosis of autism, which is currently primarily based on clinical assessments. Our approach aims to develop an early differential diagnosis based on electroencephalogram (EEG) signals, seeking to identify patterns associated with ASD. In this study, we used EEG data from 56 participants obtained from the Sheffield dataset, including 28 individuals diagnosed with Autism Spectrum Conditions (ASC) and 28 neurotypical controls, applying numerical techniques to handle missing data. Subsequently, after a detailed analysis of the signals, we applied three different starting approaches: one with the original database and the other two with selection of the most significant attributes using the PSO and evolutionary search methods. In each of these approaches, we applied a series of machine learning models, where relatively high performances for classification were observed. Results: We achieved accuracies of 99.13% ± 0.44 for the dataset with original signals, 99.23% ± 0.38 for the dataset after applying PSO, and 93.91% ± 1.10 for the dataset after the evolutionary search methodology. These results were obtained using classical classifiers, with SVM being the most effective among the first two approaches, while Random Forest with 500 trees proved more efficient in the third approach. Conclusions: Even with all the limitations of the base, the results of the experiments demonstrated promising findings in identifying patterns associated with Autism Spectrum Disorder through the analysis of EEG signals. Finally, we emphasize that this work is the starting point for a larger project with the objective of supporting and democratizing the diagnosis of ASD both in children early and later in adults.

Deficits in social interaction and communication characterize Autism Spectrum Disorder (ASD). Although widely recognized by its symptoms, diagnosing ASD remains challenging due to its wide range of clinical presentations. Methods: In this study, we propose a method to assist in the early diagnosis of autism, currently primarily based on clinical assessments. Our approach aims to develop an early differential diagnosis based on electroencephalogram (EEG) signals, seeking to identify patterns associated with ASD. We used EEG data from 56 individuals from the Sheffield dataset, applying techniques to handle missing data. Subsequently, after a detailed analysis of the signals, we applied 3 different starting approaches: one with the original database and the other two with selection of the most significant attributes using the PSO and evolutionary search methods. In each of these approaches, we applied a series of machine learning models, where relatively high performances for classification were observed. Results: We achieved an accuracy of 99.13% ± 0.44 for the dataset with original signals, 99.23% ± 0.38 for the dataset after applying PSO, and an accuracy of 93.91% ± 1.10 for the dataset after the evolutionary search methodology. These results were obtained using classical classifiers, with SVM being the most effective among the first two approaches, while Random Forest with 500 trees proved more efficient in the third approach. Conclusion: Even with all the limitations of the base, the results of the experiments demonstrated promising findings in identifying patterns associated with Autism Spectrum Disorder through the analysis of EEG signals. Finally, we emphasize that this work is the starting point for a larger project, with the objective of supporting and democratizing the diagnosis of ASD, both in children early and later in adults.

Objetivo: O Transtorno do Espectro do Autismo (TEA) é um transtorno do neurodesenvolvimento caracterizado por déficits na comunicação social e na interação social. Atualmente, o diagnóstico do TEA é clínico. Neste contexto, este trabalho busca validar uma nova proposta de ferramenta para diagnóstico diferencial do TEA, de suporte à decisão do especialista, baseada na utilização de aprendizagem de máquina e sinais de EEG. Método/abordagem: Utilizou-se o dataset 1 da base de dados Sheffield após processamento. Foram extraídos e selecionados atributos para avaliação utilizando 2 métodos: otimização por enxame de partículas e busca evolucionária. Os dois conjuntos foram divididos em treino (80%) e teste (20%), aplicou-se validação cruzada com 10 folds e, em seguida, avaliou-se 12 modelos de classificação distintos. Os experimentos foram repetidos 30 vezes. Contribuições teóricas/práticas/sociais: O modelo com melhores resultados foi o SVM Rbf 0.5, com bons valores de acurácia (99,31%), índice Kappa (0,986), sensibilidade (0,994), especificidade (0,992) e área ROC (0,993). Originalidade/relevância: Os resultados sugerem que a aprendizagem de máquina é uma ferramenta eficaz no diagnóstico do TEA.