Skip to main content

Table 2 The statistical time-domain features based on RR-interval [41]

From: An improved electrocardiogram arrhythmia classification performance with feature optimization

Statistical time-domain features

Description

Formula

mean_RR

The mean of normal-to-normal RR-intervals

\(\frac1N{\textstyle\sum_{i=1}^N}RR_i\)

std_RR

Standard deviation of RR-intervals

\(\sqrt{\frac1N}\sum\left(RR_i-{\mathrm m\mathrm e\mathrm a\mathrm n}\_\mathrm{RR}\right)^2\)

nn_50

Number of adjacent RR-intervals

\(\mathrm{Num}(\mathrm{abs}({\mathrm{RR}}_{\mathrm n+1}-{\mathrm{RR}}_{\mathrm n}))\geq50\mathrm{ms}\)

rms_RR

Root mean square of successive RR-intervals differences

\(\sqrt{\frac1{\mathrm N}}\sum{({\mathrm{RR}}_{\mathrm i+1}-{\mathrm{RR}}_{\mathrm i}})^2\)

pnn50

Percentage of NN50

\((\frac{NN50}{N})\times{100}{\%}\)

cv_RR

Coefficient of Variation of RR-interval

\(\frac{\mathrm s\mathrm{td}\_RR}{\mathrm m\mathrm e\mathrm a\mathrm{n}\_RR}\)

min_RR

Minimum RR-interval

\(\min(\mathrm{RR}-\mathrm{interval})\)

max_RR

Maximum RR-interval

\(\max(\mathrm{RR}-\mathrm{interval})\)

  1. *where RRi is i-th RR-interval in the data segment, N is total number of RR-intervals in the data segment, RRi+1 is RR-interval adjacent to RRi, and NN50 is the number obtained by dividing NN50 by the total number of NN intervals