为了更好的对致痫灶进行准确定位,提出了一种基于PCA(主成分分析)的定位方法。针对非线性动力学方法从不同角度提取癫痫脑电信号特征,首先采用主成分分析对高维特征向量进行降维处理,用随机森林进行分类;随后利用医学参考值范围找出各导联的差异变化,进而实现对致痫灶的初步定位。 In order to locate the epileptogenic zone better, a new localization method based on PCA (Principal Component Analysis) is proposed. According to the nonlinear dynamics method, the characteristics of epilepsy EEG signals were extracted from different angles. Firstly, the principal component analysis was used to reduce the dimensionality of high-dimensional eigenvectors and then classified by random forest. Then the difference of each lead was identified by using the medical reference range Change, and then to achieve the initial location of the epileptogenic focus.