Introduction

Time-Domain Analysis

The time-domain analysis contains the following analysis methods:

Mean:

This is the mean value of the RR intervals. It is calculated by summing all NN intervals and then dividing by their number. Read more

SDNN:

This is the standard deviation of the NN intervals. Read more

RMSSD:

This is the root mean square of the differences between successive NN intervals. Read more

SDSD:

This is the standard deviation of the differences between successive NN intervals. Read more

NN20/NN50:

This is the number of pairs of successive NN intervals that differ by more than 20ms/50ms. Read more

pNN20/pNN50:

This is the percentage of pairs of successive NN intervals that differ by more than 20ms/50ms. Read more

rRR:

The relative RR intervals are calculated using the equation
for i=2...n

\[rr _{i} := \frac{2*(RR_{i}-RR_{i-1})}{RR_{i}+RR_{i-1}}\]

where n is the number of RR intervals.
The HRV is measured by the median of the euclidean distances of the relative RR intervals to the average of the relative RR intervals. Read more

Frequency-Domain Analysis

Frequency domain analysis uses a Lomb Scargle Transformation to determine the power spectral density of each frequency domain. The frequency bands are defined as follows:

VLF: very low frequency, from 0.003 to 0.04 Hz
LF: low frequency, from 0.04 to 0.15 Hz
HF: high frequency, from 0.15 to 0.4 Hz
LF/HF: The ratio of LF and HF
Total Power: The sum of VLF, LF and HF

Geometric Analysis

Poincaré plot

This plot is used to quantify self-similarity in processes. Read more

Recurrence plot

This plot is used to visualize the periodic nature of a trajectory through a phase space. Read more