Study On Electroencephalography Preprocessing Methods In Non-visual Biological Effects

Non-visual biological effects are mainly involved in the regulation of human biological rhythms,usually by observing melatonin secretion and pupil size.However,they does not specifically reflect the effects of non-visual effects on brain function.Human brain electrical signals contain rich information,can reflect the activities of brain function,and study non-visual biological effects through EEG signals.It is an objective and effective method,and has become a research hotspot in the field of non-visual biological effects at home and abroad.However,EEG signals are easily interfered with by various physiological and non-physiological artifacts.Traditional or single pretreatment methods cannot meet the requirements of EEG pretreatment in non-visual biological effects.This paper first proposes a hypothesis that if the blue light with a flashing frequency of 40 Hz is repeatedly stimulated to the human eye,the activity of certain regions of the brain may be enhanced or weakened.The response on the EEG is corresponding to the gamma rhythm corresponding to the scintillation frequency.change.Then,in order to remove the artifacts in the EEG signal,two methods of artifact removal are proposed in this paper.The effectiveness and practical value of the two preprocessing methods are verified by simulation and actual acquisition of EEG.Finally,the design experiment validated the hypothesis.The main research work of this paper is as follows:1)A denoising method for improving the soft and hard threshold is proposed.Firstly,the continuity and asymptotic line of the threshold function are analyzed in detail.Then,in order to illustrate the effectiveness of the method,the international standard EEG data simulation experiment and similar methods are adopted.Through the calculation of signal-to-noise ratio and root mean square error,it is found that this method has improved on these two indicators compared with other methods.Finally,the method is used in the actual collected EEG data,and the results show that it can retain more detailed information and better smoothness.2)A denoising method combining discrete wavelet and FastICA is proposed.Firstly,the steps of removing artifacts are analyzed in detail.Then,the effectiveness of this method is verified by the international standard EEG simulation experiment.The results and other methods are used.In comparison,there is an increase of 1.5% to 16.4% in the correlation coefficient.Finally,the method and the same method are applied to the actually collected EEG signals.It is also found that the method can effectively remove artifacts and ensure the integrity of real EEG signals.3)In order to verify the proposed hypothesis,the laboratory designed a blue-light irritating glasses to collect experimental data of 35 elderly people.Actually,20 people were selected as the analysis object,and the Simple Mental State Scale(MMSE)score was used as the grouping standard.The EEG signal of 20 elderly people was preprocessed by a pre-processing method combining discrete wavelet and FastICA,and the Butterworth high-pass filter was used to extract the ? rhythm.By calculating the power spectrum value of ? rhythm and comparing the differences between different groups,it is verified that the assumption is correct.In general,the two EEG signal preprocessing methods proposed in this paper can effectively remove artifacts in EEG signals and lay a good foundation for further analysis and processing of EEG signals.At the same time,verification experiments have also shown that non-visual biological effects can indeed affect brain function,and biological rhythms can be adjusted by external stimulation.