EEG classification for acupuncture manipulation based on wavelet transform and random forest algorithm
ZHANG Ke
YAO Yong
XU Gang
YANG Huayuan
Abstract：To investigate the specificity of different acupuncture manipulations on electroencephalogram(EEG)signals,the EEG signals after acupuncture manipulations at Quchi(LI11)were classified based on wavelet transform and random forest algorithm.First-ly,18 healthy volunteers were performed three kinds of acupuncture intervention at LI11(reinforcing,reducing,and even reinforcing-re-ducing methods).The scalp EEG signals of 100 s during acupuncture were recorded and segmented by 2 s.Secondly,the mean energy and wavelet entropy of δ,θ,α,β and γ waves were extracted by wavelet transform to construct the feature set.Finally,the features were sorted by maximal-relevance-minimal-redundancy algorithm,combined with the random forest algorithm,and the model perform-ance assessed via 10-fold cross-validation.The results showed that the classification accuracy of the model based on random forest al-gorithm could reach 0.950 for the EEG signals of three kinds of acupuncture manipulations.The research enables the classification of EEG signals induced by three acupuncture manipulations at LI11,indicating the significant differences of acupuncture manipulations in the regulation of brain function,and can provide a new method and approach for the quantitative evaluation of acupuncture efficacy.
Keywords：Acupuncture manipulationsEletroencephalogramWavelet transformRandom forestQuchi acupoint
Publication Date：2025-02-28
Online Publishing Date：2026-09-11(First online date of this platform, not the publication date of the document)
Pages：7( 1-7 )