Individual identification method for specificity of cortex electroencephalogram signals in operation

Abstract

The invention discloses an individual identification method for the specificity of cortex electroencephalogram signals in operation. The method comprises the following steps: 1, collecting ECoGs of four points of an individual movement area, a language area, a sensory area and a non-functional area determined through ECS, and ECoGs leading to the operative region 64 / 128 of the same individual, and carrying out pretreatment and feature extraction on the ECoGs, thus obtaining feature samples; 2, carrying out model training and optimizing by utilizing the feature samples of the ECoGs of the four points determined through ECS, and thus the individual optimal identification model is trained; and 3, classifying the feature samples of the ECoGs leading to the operative region 64 / 128 of the same individual by utilizing the individual optimal recognition model trained in the step 2, so that the attributes of all the lead signals are obtained, and the specific recognition for the ECoGs of each brain function region of the individual is completed. With the adoption of the method, the ECOGs of different functional states can be quickly, accurately and comprehensively identified, so that powerful assistance is provided for identifying the electroencephalogram activities in different functional states.

Description

technical field [0001] The invention relates to the field of biomedical signal processing, in particular to an individualized identification method specific to intraoperative cortical EEG signals. Background technique [0002] As a means to measure the intensity of brain nerve activity, EEG is widely used in the diagnosis of epilepsy and other neurological diseases due to its advantages of non-invasiveness, low cost, and convenience. At the same time, as an extremely weak signal, the EEG signal is easily interfered by internal signals such as myoelectricity and oculoelectricity, as well as external signals such as power frequency interference and electromagnetic interference. In particular, scalp EEG signals are more susceptible to interference because the electrodes are blocked by the skull and scalp, so the present invention uses cortical EEG signals collected after craniotomy with less interference. According to domestic and foreign literature reports, when the brain is ...

AI Technical Summary

Problems solved by technology

[0003] Due to the differences in the EEG of different individuals, the difference in the recognition model of the cortical ECoG signal (ECoG) is prone to the problem of low accuracy of allogeneic detection

Images