Method for identifying and quantifying predictive geography characterization of brain spontaneous activity

The invention provides a method for identifying and quantifying predictive pattern characterization of brain spontaneous activity. Comprising the following steps: based on functional magnetic resonance image data in a resting state, adopting a time-varying dynamic analysis method to identify a dynamic state of brain spontaneous activity, and based on mode similarity analysis and a subsequent representation model, constructing a predictive representation map M, and meanwhile, adopting dimensionality reduction and clustering algorithms to obtain a two-dimensional space of representation distribution in a target state, mapping the predictive characterization map M into the space to realize visualization of the perception domain of the target state; and detecting and quantifying predictive offset in the internal transfer process of the target state by using two mathematical measurement design algorithms of kurtosis and skewness to obtain a predictive offset value of the brain spontaneous activity at an individual