Discriminant Analysis in Schizophrenia based on Neural Network

In this paper, we use neural network to classify schizophreniapatients and healthy control subjects. Based on 4005 high dimensions feature space consistof functional connectivity about 63 schizophrenic patients and 57 healthy control as theoriginal data, attempting to try different dimensionality reduction methods, differentneural network model to find the optimal classification model. The results show that usingthe Mann-Whitney U test to select the more discrimination features as input and usingElman neural network model for classification to get the best results, can reach a highestaccuracy of 94.17%, with the sensitivity being 92.06% and the specificity being 96.49%.For the best classification neural network model, we identified 34 consensus functionalconnectivities that exhibit high discriminative power in classification, which includes 26brain regions, particularly in the thalamus regions corresponding to the maximum number offunctional connectivity edges, followed by the cingulate gyrus and frontal region.