A method for eliminating myoelectric artifacts in state-related dynamic EEG signals

Abstract

The invention discloses a method for eliminating myoelectric artifacts in state-related dynamic EEG signals. The steps include: 1. First, construct two data sets by delaying the collected EEG observation signals; 2. Using this The hidden Markov independent vector analysis method proposed by the invention performs dynamic joint blind source separation to obtain the source signal matrix and unmixing matrix of each data set in each state; 3. Select the source signal matrix and unmixing matrix corresponding to the EEG signal 4. Sorting each independent source component in the source signal matrix according to the autocorrelation coefficient, selecting the independent source component related to the EMG noise and setting it to zero; 5. Blind source separation and inverse transformation to obtain a clean EEG signal after noise elimination. The present invention can remove the influence of myoelectric noise on EEG signals in an actual dynamic environment, and at the same time keep the information of EEG activity as much as possible without loss, thereby improving the analysis accuracy of EEG signals and providing a solution for EEG denoising. A new way of thinking.

Description

technical field [0001] The invention relates to the field of medical signal processing, in particular to a method for removing noise from neural signals in complex environments. Background technique [0002] By measuring the electrical signals generated by neuron activity, electroencephalography (EEG) can non-invasively and conveniently observe and record brain activity, and has been widely used in neuroscience research, disease diagnosis and medical health monitoring. However, due to the weak amplitude of EEG signals, they are often contaminated by various non-neurophysiological factors originating from eye, heart and muscle activity. In actual long-term medical monitoring, these unavoidable noises will obviously interfere with the recorded EEG signals, thus adversely affecting the subsequent analysis. Therefore, it is of great significance to develop effective noise removal algorithms without affecting the real EEG data. Compared with oculoelectric and electrocardiograph...

AI Technical Summary

Problems solved by technology

However, the existing EEG signal denoising methods are mostly suitable for static situations, that is, the model assumes that the mixing mode of EEG recordings does not change during the entire observation process.

When the traditional static EEG noise reduction method is applied in the actual mobile medical monitoring, the EMG noise cannot be effectively removed, which will cause serious interference to the subsequent research on EEG recording.

Images