Hi, I have read about AEC methods from this paper:
https://pubmed.ncbi.nlm.nih.gov/21352925/
- They describe 2 methods bellow, I wonder which one is used in Brainstorm?
- During AEC computation continuous data is divided into epochs, what is the default length of the epoch in Brainstorm?
- In order to remove linear dependencies from the signal and control for source leakage, did you apply orthogonalization in Brainstorm?
-
AEC: The timecourses of bandpass filtered electrical activity at the
seed and test voxels were Hilbert transformed to obtain the
analytic signal. The absolute value of the analytic signal was then
determined in order to give the envelope of oscillatory power in
the frequency band of interest. This we term the Hilbert envelope
and this technique has been used extensively in previous MEG
studies (for a mathematical description see Brookes et al., 2004).
The Hilbert envelopes for the seed and test voxels were divided
into n time segments of equal length Δ. The Pearson correlation
coefficient between seed and test Hilbert envelopes was computed
within each segment. This gave n correlation values, one for each
time segment. These were then averaged across segments yielding
a single average value which we term Averaged Envelope
Correlation (AEC). -
CAE: The timecourses of electrical activity at both the seed and test
voxels were Hilbert transformed and the Hilbert envelopes computed.
The Hilbert envelopes were divided into n time segments of
equal length Δ and the average value of the envelope computed
within each time window. This resulted in two new temporally
averaged Hilbert envelope timecourses, each comprising n points
(we term these the ‘averaged envelopes’). The Pearson correlation
between these averaged envelopeswas then computed as ameasure
of FC which we term Correlation of Average Envelopes (CAE).