ECoG intraoperative brain function positioning method based on clustering and classification algorithms

Abstract

The invention discloses an ECoG intraoperative brain function positioning method based on clustering and classification algorithms. The method comprises the following steps: firstly, the energy proportion of six layers of wavelet decomposition reconstruction single sub-bands is extracted as a characteristic quantity by employing the functional region boundary clustering algorithm, and the class clusters of all leads are clustered by employing a cohesion-type hierarchical clustering model; secondly, a functional area attribute classification algorithm is adopted, six-dimensional time domain statistics and five-dimensional rhythm energy are used as characteristic quantities, PCA is used for carrying out dimensionality reduction on characteristics, and SVM is used for classifying lead functional area attributes; and finally, the algorithms are integrated to establish an ECoG intraoperative brain function positioning integration algorithm, and the clustered class clusters are matched with the functional area attributes by adopting a self-defined similarity algorithm to complete functional area positioning. The algorithm model has good generalization, can realize accurate and rapid intraoperative brain function positioning clinical application, and can be widely applied to neuroscience researches such as intraoperative brain function positioning based on electroencephalogram analysis.

Description

technical field [0001] The invention relates to the research field of neuroscience research and clinical application of electroencephalogram analysis, in particular to a brain function localization method in ECoG operation based on clustering and classification algorithms. Background technique [0002] The brain is the center of the human nervous system. Brain lesions will affect human memory, language, movement and other functions, seriously affecting human survival and quality of life. Among them, epilepsy and glioma are the most common brain lesions one. At present, the main treatment for such brain lesions is neurosurgery, and the most important step in neurosurgery is intraoperative brain function localization. Traditional intraoperative brain function localization methods (anatomy, image, cortical electrical stimulation, etc.) have problems of speed, accuracy, and possible trauma. The intraoperative brain function localization method using cortical electroencephalogra...

AI Technical Summary

Problems solved by technology

However, the low accuracy rate of allogeneic detection indicates that there is no common specificity in the brain functional areas of the population, or it is very difficult to find out the common specificity

Images