Usage: findPhysioPeaks fidDir | {type file}
where: type = {acq | respir | cardio }
file = {sdtRoot | name.asc}
There are two ways to specifiy which files to process
on the command line. The first method is to specify
only the fid directory,
fidDir.
example> findPhysioPeaks subject001.fid
It is then assumed that
the following three ascii files exist in this directory:
(Note that an optional filename extension is allowed).
- ch0 - respiration ascii file
- ch1 - cardiac ascii file
- ch2 - acquisition pulse ascii file
The second method is to specify each of the files individually
along with the type of data it contains.
If no file extension is given, the data is assumed to
be an sdt/spr file pair, otherwise it is assumed to
be an ascii file.
The
.sdt files must be 1D Real (i.e. float)
and the
.asc must be ascii with one value
per line.
See
Sdt File Format for details.
For example:
example> findPhysioPeaks acq acqData.asc
example> findPhysioPeaks respir respirData.asc
example> findPhysioPeaks cardio cardioData.asc
Prior to determining the peak locations, the input
waveforms are conditioned. The
acq waveform is median
filtered to remove any transient spikes. The
respir and
cardio waveforms are bandpass filtered via convolution to
remove DC drift and high frequency noise. The
cardio
waveform is also inverted to make the "S" dip of the QRS
complex the peak since it is the most pronounced feature.
For each of the three input files a pair of output sdt files
are generated; a conditioned waveform and a peak waveform.
- The peak sdt contains the value of 1.0 at each peak
of every cycle in the Respiration/Cardiac waveform.
For the acquisition waveform, a value of 1.0 is set at
the location of each rising and falling edge for the pulse.
Non-peak or Non-edge locations have a value of 0.0.
- The cond sdt contains the resultant waveform after
the specified processing.
NOTE:
The current implementation of this conditioning
requires a 100 Hz sampling rate.
