Anslab crossspectral
analysis allows the extraction of crossspectral parameters from two
signal previously extracted using other anslab analyses like ecg, resp
or bp. One possible application could be to explore crossspectral
features of breathing pattern and heart rate to quantify respiratory
sinus arhythmia. For each of the two signals, the spectral
density is calculated and the cross spectral density of the two.
The crossspectrum is used to identify the main shared frequency of the
two signals in a user specified band (VLF, LF or HF). For this
frequency, the transfer function, the coherence and the crossspectral
phase angle is extracted, as shown on the example below.
You can specifiy the signals to be used, detrending options, band of interest and other options on the crossspectral optionsdialog shown below. The example below uses the traces RS0 (tidal volume) from the respiration analysis and the ibi0-trace from the ecg analysis. Like the spectral analysis, you start by selecting the raw (!) data file you wish to process. Anslab then looks for the specified variables in the dependant result data file in the given subfolders of the specified channel type subfolders of the study folder (so for the example below, anslab will try to find the RS0variable in a data file taken from the 'RESP-1abc'-subfolder in the 'resp'-subfolder and the ibi0-variable in a data file taken from the 'ECG Lead II'-subfolder in the ecg-subfolder of the selected study folder). Anslab assumes that these signals are sampled with the here given sampling rates (both RS0 and ibi0 -variable are saved with 4 Hz by default).
timing files
Before loading the data, anslab will check if a spectral-timingfile is
present next to the raw datafile. If so, intervals to be
processed will
be taken from that file. A spectral timing file has
the same structure as standard anslab timing files, except that it's
named 'MyFileName.spectral.m'
instead of 'MyFileName.m' (see 
timing
files for more information). You can create such a timing file by
running marker
analysis. You can additionally subdivide intervals created with the
marker analysis using timing file modification
from the tools menu. Note that the timing file for crossspectral and
spectral analyses are the same, as it is assumed that if both analyses
are run, they will in almost all cases be run for the same intervals.
Anslab then loads the signals, jumps to the first interval defined,
and
displays the signals, an underlying trend and the resulting power
spectra, crossspectrum, transfer function, coherence and phase spectrum
for this piece of data. RSA frequency bands
are colorized, and a local maximum of the crossspectrum is identified,
which is used for the extraction of transfer function, phase angle and
coherence values. This maximum is plotted as a red line in the
crossspectrum, and you can drag this line to an optimal position. The
band wherein the maximum is automatically searched can be set on the
options dialog using the 'cross spectral range of
interest'-dropdownlist. Note that when releasing the red line at
a new position, tfe, coh and pha-values are updated. As for the
spectral analysis, you can also adjust the window borders, by draging
the edges of the grey surfaces to enclose only an interval of good
data. When you're done, hit the accept button on the dynamic section
(shown below) to go to the next segment defined in the timing
file. If an entire interval cannot be used, hit the set missing button
(tfe, coh and pha are displayes as NaN) and hit accept to save the
missing values and jump to the next segment. The
<<<-button allows you to go 1 segment backwards, the restart
button will erase all editing and begin with segment 1, and the clear
segment-button will only erase editing in the current segment. Using
the jump-to-segment-editbox, you can jump to a segment of your choice.
