Denoise EEG signal by using Daubechies function

Question

I have an EEG signal and it contains eye blink artifacts, i read some references and know that can detect eye blink and remove them by using wavelet transform but i don't know that how do it, How to detect eye blink? Have any tutorials for me, after transformed EEG signal into wavelet coefficients, what should i do and which level of daubechies can be used to do that? Thank you!

Answer 1

I don't know whether this will work but you can give it a try.

Wavelet transform works like a filter bank. Set the wavelet level to such a value so that the last level of the decomposition gives you a filter bank of nearly 0Hz - 5Hz.

Get the coefficients of the detail functions at this level and do a thresholding (soft/hard) on the same.and then compose back the signal using the new coeffiecients Blinks have a relatively high amplitude and thresholding on them might give you what you want.

Answer 2

If you want to remove eye blinks, a commonly used approach is running Independent Component Analysis (ICA) on the data, identifying the blink artifact and backtransforming to the original data without that independent component. There are other approaches, but ICA works quite well even in very noisy EEG data (e.g. from simultaneous EEG-fMRI sessions).

Answer 3

Eye blinks will generally have a frequency between 2-5Hz.

You can first train a system to capture eyeblinks. Then use the same to detect the blinks in an eeg signal