Calculate epochs statistics (mean, median, standard deviation, variance, kurtosis, skewness, mean diff value, median diff value, max difference, deviation from channel mean):
epoch_stats(eeg)Calculate channels statistics (mean, median, standard deviation, variance, kurtosis, skewness, mean diff value, median diff value, max difference, deviation from channel mean) per epoch:
channel_stats(eeg)Calculate covariance matrix:
covm(eeg)Calculate correlation matrix
corm(eeg)Calculate auto-covariance:
acov(eeg, lag=20, norm=false)Calculate cross-covariance:
cc, lags = xcov(eeg, lag=20, demean=true)
eeg1 = filter(eeg, fprototype=:butterworth, ftype=:bs, cutoff=(45, 55), order=8)
eeg2 = filter(eeg, fprototype=:butterworth, ftype=:bs, cutoff=(45, 55), order=12)
cc, lags = xcov(eeg1, eeg2, ch1="F3", ch2="F4", ep1=1, ep2=1, lag=20, demean=true, norm=true)Mutual information:
m = mutual_information(eeg, ch=get_channel(eeg, type="eeg"))
plot_matrix(m[:, :, 1], xlabels=labels(eeg)[1:19], ylabels=labels(eeg)[1:19])(!) Currently only one estimator (maximum likelihood) is available.
Internally it uses
InformationMeasures.get_mutual_information(), source: InformationMeasures.jl.
Entropy:
e, _, _ = entropy(eeg, ch=get_channel(eeg, type="eeg"))
plot_bar(e[:, 1], labels=labels(eeg)[1:19], seriestype=:bar, title="Entropy, epoch 1")(!) Entropy of the signal is calculated using its histogram bins
(number of bins is calculated using the Freedman-Diaconis formula) using
the formulas p = n / sum(n) and
entropy = -sum(p .* log2(p)), where p is the
probability of each bin and n are bins’ weights. Shannon
entropy and log energy entropy are calculated using
Wavelets.coefentropy().
Calculate ISPC (Inter-Site-Phase Clustering):
ispc(eeg, eeg, ch1=get_channel(eeg, type="eeg")[1:5], ch2=get_channel(eeg, type="eeg")[6:10], ep1=1, ep2=1)Calculate PLI (Phase Lag Index):
pli(eeg, eeg, ch1="F3", ch2="F4", ep1=1, ep2=1)Amplitude Envelope Correlation:
env_cor(eeg, eeg, ch1="F3", ch2="F4", ep1=1, ep2=2)Perform discrete wavelet transform (DWT) and continuous wavelet transform (CWT):
dw_trans(e10, wt=wavelet(WT.haar), type=:sdwt)
cw_trans(e10, wt=wavelet(Morlet(π), β=2))SNR (signal-to-noise) may be estimated from epoched EEG signal.
snr, hz = snr(eeg)
p = plot(hz, snr[1, :], label="ch 1", xlabel="Frequency [Hz]", ylabel="SNR")
p = plot!(hz, snr[2, :], label="ch 2")
p = plot!(hz, snr[3, :], label="ch 3")
p = plot!(hz, snr[4, :], label="ch 4")
plot_save(p, file_name="images/eeg_snr.png")