Research on recognition method of adolescent schizophrenia based on EEG feature fusion

Schizophrenia is a common mental illness, and adolescent schizophrenia has a higher recurrence rate and disability rate than adult schizophrenia. In order to improve the recognition accuracy of adolescent schizophrenia, this study proposed a classification method of adolescent schizophrenia combined with artificial intelligence technology. In this method, the pre-processed original electroencephalogram (EEG) signals are decomposed by Discrete Wavelet Transform (DWT) and Ensemble Empirical Mode Decomposition (EEMD), respectively, and then the decomposed signals are extracted and the extracted features are fused. Finally, five machine learning models are used for classification. The results show that compared with DWT or EEMD methods alone, the feature fusion method proposed in this paper exhibits a higher accuracy in classification, in which the K-Nearest Neighbours (KNN) model achieves the best classification performance. The identification accuracy, recall, precision, F1 score, and Kappa coefficient of the model were 97.48 ^% , 95.71%, 99.56%, 97.59%, and 0.9494, respectively, which proved that the proposed method could effectively distinguish adolescent schizophrenia patients from healthy adolescents, and could assist doctors in disease analysis, providing a new method for the diagnosis of adolescent schizophrenia.