Neurological disease identification method based on causal-time delay brain network

The invention discloses a neurological disease identification method based on a causal-time delay brain network, and the method comprises the steps: firstly obtaining a brain function image of a subject, and extracting time sequence data; secondly, performing normalization operation on the time sequence data according to columns, and obtaining a first-order difference matrix of the time sequence data; then, a generative adversarial network is constructed, and the generative adversarial network comprises two gating structures used for estimating the causal effect and the time-delay effect of the brain interval; then, when the model is trained, time sequence data and a first-order difference matrix of the time sequence data are used as inputs of the generative adversarial network at the same time; and finally, features are extracted based on the constructed brain causal and time-delay network, and a classifier is trained to realize diagnosis of neurological diseases. Estimation of the causal time-delay effect i