Identification method for EEG (electroencephalogram) signals in different physiological states based on neighbor visual length entropies

Abstract

An identification method for EEG signals in different physiological states based on neighbor visual length entropies comprises that original EEG signal data is obtained, the EEG signal data is drawn in a coordinate axis in equal horizontal lines in the form of histogram; three data points closes to a data point are defined as neighbors in first, second and third levels respectively, and the visualrelations between the data point and the neighbors in first, second and third levels are determined; the visual lengths between the data point and the neighbors satisfying visual conditions are obtained respectively; the visual length entropies of the neighbors in first, second and third levels are calculated respectively; a combined distribution map of all EEG signals is drawn according to the visual length entropies of the neighbors in first, second and third levels; and according to the visual length entropies of the neighbors in first, second and third levels, 10-fold cross-validation iscombined with a soft-interval support vector machine to classify the EEC signals. The leveled neighbor visual length entropies of visual degree differences between the significant neighbors and the data point are taken into integrated consideration, and the EEG signals of different physiological states are classified accurately.

Description

technical field [0001] The invention relates to a recognition method of electroencephalogram signals. In particular, it relates to a method for identifying EEG signals in different physiological states based on the neighbor visual length entropy of each data point in a time series. Background technique [0002] EEG signals are the overall reflection of the electrophysiological activities of brain nerve cells on the surface of the cerebral cortex or scalp. When the human body is in the same physiological state (such as a certain stage of sleep: light sleep stage), the EEG signal will show similar fluctuations, and when the physiological state changes (such as from light sleep stage to deep sleep stage), its Significant changes will also occur in EEG signals. Accurate classification of EEG signals under different physiological states can help us better understand the working mechanism of the brain and promote the further development of brain science research. However, EEG s...

AI Technical Summary

Problems solved by technology

It is difficult to obtain satisfactory results when the above method is used for complex time series analysis

Images