[ top | up ]

Approximate Entropy (ApEn) of RR or HR

Syntax

[ntimes,vals,labs] = apenhr(times,vals,labels,[rstd lag]);

Arguments

times A column vector of times at which each val is measured.
vals A column vector of data.
labels A column vector of labels indicated whether the data are valid. 1 indicates valid data, 0 indicates invalid data.
rstd The "filter factor" in ApEn. If this is positive, it will be multiplied by the standard deviation of vals (after invalid points are deleted) to set the ApEn filter factor. If rstd is negative, its absolute value will be used directly as the ApEn filter factor. The default value is 0.2 (which will be multiplied by the standard deviation of the data).
lag The embedding lag. In Kaplan et al., this is set to one beat.

Returned Values

ntimes The min and max time of the segment being analyzed.
nval The approximate entropy of this segment.
labels The fraction of the data that was valid.

Description

apenhr implements the ApEn calculation for heart rate, using the standard parameter values: an embedding dimension of 2 and an embedding lag of 1 sample. (If heart rate is being analyzed, the literature suggests that it should be arranged to have a sampling interval of roughly 0.8 secs.) The program carries out the embedding steps etc. detailed in the documentation for apen.

Invalid data are eliminated from the calculation after the embedding has been performed, so that only completely valid vectors are used.

References

See Also

apen.

Examples