EEG signals combined with feature fusion in the diagnoses of schizophrenia and depression
Objective To explore machine learning algorithms combined with EEG signals in the diagnoses of schizophrenia and depression.Methods EEG signals from 31 patients with schizophrenia and 28 patients with depression were selected and converted from EDF format to ASCII format.Features such as Lempel-Ziv complexity,maximum Lyapunov exponent,and Higuchi fractal dimension were extracted from the EEG signals.These features were then fused using a feature fusion strategy to create new feature vectors,which were classified through machine learning algorithms.Results The classification accuracy of the support vector machine(SVM)with a Gaussian kernel was 84.85％,with a sensitivity of 89.47％and a specificity of 78.57％.Conclusion This study demonstrates the potential of combining EEG signal feature extraction with machine learning algorithms for the identification of schizophrenia and depression,offering valuable insights for developing novel diagnostic techniques for these mental disorders.
NETL
NSTL
万方数据
