传递熵
多元统计
熵(时间箭头)
稳健性(进化)
注意缺陷多动障碍
样本熵
数学
传递函数
二进制数
人工智能
计算机科学
多元微积分
色散(光学)
模式识别(心理学)
广义熵指数
多元分析
统计物理学
统计
信息传递
复杂系统
算法
分散指数
近似熵
机器学习
生物系统
信息论
二进制数据
标识
DOI:10.1142/s0219477526500392
摘要
In this paper, we propose multivariate multiscale weighted dispersion transfer entropy (MM-WDTE). This method introduces weight coefficients and weight probabilities in the symbolization process of dispersion patterns to retain key information to the greatest extent and effectively improve the robustness to noise. Furthermore, this method has been extended to cover multivariable systems and multiscale levels, breaking through the limitations of the traditional binary transfer entropy model on dimensions, and comprehensively grasping the dynamic characteristics of signals on multiple time scales. By applying this method, we performed an interaction analysis of multi-channel electroencephalogram (EEG) data in four classical frequency bands (Theta, Alpha, Beta, Gamma) in children with attention deficit hyperactivity disorder (ADHD). Based on the positions of the electrodes, the brain was separated into four regions. By using a cross-regional study, we identified differences in information interaction between different brain regions in children with ADHD and controls.
科研通智能强力驱动
Strongly Powered by AbleSci AI