# Sensor Systems

### Sidebar

methods:hrv:hrv_main

# Heart Rate Variability

## What is Heart Rate Variability?

This introduction is adapted from an earlier publication1)

The heart works as an autonomic oscillator, it is not triggered by the brain or something else. However, the autonomic nervous system controls the activity of the heart indirectly through a number of physiological pathways. Next to hormones, such as adrenalin (epinephrine), there are two nerve-based pathways. The first is the sympathetic branch of the autonomic nervous system (sympaticus), which has the effect of increasing heart rate and force of the heart beats. The second is the parasympathetic branch of the autonomic nervous system (parasympaticus), which acts as a kind of brake. Normally both branches are active simultaneously and form an equilibrium. These pathways are part of a feedback control system, the main function of which is to adjust the blood flow to the needs and the anticipated needs of the body while keeping the blood pressure within safe limits. In the field of physiology this concept is known as homeostasis: keeping physiological systems constant, or at least within bounds. Examples of important physiological systems that are kept constant are body temperature ($37\,^{\circ}\mathrm{C}$), salt level ($0.9%$ NaCl), sugar level ($0.1%$ glucose), and blood volume ($5$-$6$ liters). The body has an extensive networks of sensors and effectors to maintain homeostasis.

The heart rate control loop keeps the blood pressure within safe limits. In such feedback loops, there are delays (for example due to the slow neurotransmitters of the sympaticus) and storage capacities (the elastic buffer capacity of the big arterial vessels). As a result, the heart rate is not completely stable (not even when the person is at rest), nor does it enter an unbounded oscillation. The variation of heart rate is known as HRV (Heart Rate Variability). A non-zero HRV is a sign that the feedback loop is working. The statistics and the frequency components of HRV are studied intensively in the context of stress measurement. For certain types of patients, low HRV can be correlated with certain heart- and blood-vessel related disorders. There are many options for measures to indicate the degree of HRV, the trick is in picking the righ one. There are tools such as Kubios to calculate a wide variety of HRV measures2).

An interesting phenomenon in the dynamics of HRV is that the system tends to have an eigen-frequency near $0.1$ cycles per second. The precise value differs somewhat from person to person. Another interesting phenomenon is that under conditions of stress, the HRV tends to be lower than under normal conditions when the person is relaxed. This can be explained by the fact that under stress the parasympatic brake function is so diminished, that it is not regulating anymore (note that this is the regulator with the fast neurotransmitters responsible for most of the fast dynamics). The mechanisms are by now fairly well understood, see for example DeBoer3).

This page gives some links to basic formulas for implementing heart-rate and heart-rate variability algorithms in your software when using any method for measuring heart rate (optical (PPG) or electrical (ECG)). It also provides some terminology commonly used for stochastic signals. The input signal is the digital interval information from the Arduino, which is asynchronous. In an asynchronous signal, the samples are not equi-distant in time.

## HRV standards

Some representations of the average, standard deviation and frequency components have become widely acceptable in the field of heart-rate variability4) 5).

## Time-domain methods

These are based on the beat-to-beat or NN intervals, which are analyzed to give variables such as:

• SDNN, the standard deviation of NN intervals. Often calculated over a 24-hour period.
• SDANN, the standard deviation of the average NN intervals calculated over short periods, usually 5 minutes. SDANN is therefore a measure of changes in heart rate due to cycles longer than 5 minutes.
• RMSSD, the square root of the mean squared difference of successive NNs.
• NN50, the number of pairs of successive NNs that differ by more than $50 ms$.
• pNN50, the proportion of NN50 divided by total number of NNs.

Geometric methods are a subclass of time-domain measures in which the NN intervals are converted to a geometric pattern, then analyzed. Most of these methods use a discrete scale for the NN interval length. In a Poincaré plot, the relation between two neighbouring intervals is plotted as shown in figure 1. It is described scientifically and mathematically correct by Brennan in 20016), a paper which is also useful for the other time-domain evaluation methods.

Fig. 1: Poincaré plot with on the horizontal axis the NN intervals and on the vertical axis the length of the previous NN interval

These spatial time plots appear to be very suitable for direct bio-feedback methods when constructed on the fly7) 8)

## Frequency-domain methods

Several methods are available. Power spectral density, using parametric or nonparametric methods, provides basic information of the power (variation) distribution across frequencies. One of the most commonly used PSD methods is the Fast Fourier transform. Several frequency bands of interest have been defined in humans:

• High Frequency band (HF) between $0.15$ and $0.4 Hz$. HF is driven by respiration and appears to derive mainly from vagal activity (parasympathetic nervous system).
• Low Frequency band (LF) between $0.04$ and $0.15 Hz$. LF derives from both parasympathetic and sympathetic activity and has been hypothesized to reflect the delay in the baroreceptor loop.
• Very Low Frequency band (VLF) band between $0.0033$ and $0.04 Hz$. The origin of VLF is not well known.
• Ultra Low Frequency (ULF) band between $0$ and $0.0033 Hz$. The major background of ULF is day–night variation and therefore is only expressed in 24-hour recordings.

The ratio of low-to-high frequency spectra power (LF/HF) is has been proposed as an index of sympathetic to parasympathetic balance of heart rate fluctuation, but this is controversial because of the lack of understanding of the mechanisms for the LF component.

A good course on Signal Processing is given by William D Penny of UCL London9).

1) , 8)
Loe Feijs, Geert van Boxtel, Geert Langereis, Designing for the Movements of Heart Rate and Breath, DeSForM 2010, 6th International Workshop on Design & Semantics of Form & Movement: Design Semantics in Context, November 3-5 2010, Lucerne, Switzerland
2)
Tarvainen M. & Niskanen JP. Kubios HRV Analysis. Version 2.0 beta. User's guide. Department of Physics. University of Kuopio. http://bsamig.uku.fi/kubios/kubios_hrv_users_guide.pdf (2.4.2008)
3)
DeBoer RW, Karemaker JM & Strackee J. Hemodynamic fluctuations and baroreflex sensitivity in humans: a beat-to-beat model. Am J Physiol Heart Circ Physiol 253: H680–H689, 1987.
4)
M. Malik, J.T. Bigger, A.J. Camm, R.E. Kleiger, Heart rate variability: Standards of measurement, physiological interpretation, European heart journal, 1996, volume: 17, issue: 3, pp. 354-381
5)
Wikipedia, page on Heart Rate Variability, http://en.wikipedia.org/wiki/Heart_rate_variability
6)
M. Brennan, M. Palaniswami, P. Kamen, Do existing measures of Poincare plot geometry reflect nonlinear features of heart rate variability?, IEEE Transactions on Biomedical Engineering, Volume 48, Issue 11, Nov. 2001, pp. 1342 – 1347
7)
Loe Feijs, Geert Langereis and Geert Van Boxtel, Alternative Presentations of HRV Feedback, In: Abstracts of Scientific Papers Presented at the 15th Annual Meeting of the Biofeedback Foundation of Europe, Appl Psychophysiol Biofeedback (2011) 36:p. 290.
9)
William D Penny, Signal Processing Course, UCL London, 1999, http://www.fil.ion.ucl.ac.uk/~wpenny/course/course.html