Acquire ECG
Record a signal with sufficient electrode contact and minimal artefact.
Scientific guide · ECG · RR · HRV
ECG & HRV analysis: from raw signal to meaningful trends.
Heart rate variability begins with timing, but reliable interpretation begins earlier — with the quality of the recorded signal and correct identification of individual beats. This guide explains how ECG, R peaks, RR/NN intervals and HRV metrics fit together, and why measurement conditions matter when recordings are compared over time.
ECGElectrical waveform and beat morphology
R peaksTiming reference for beat detection
RR / NNBeat-to-beat interval sequence
HRVVariability described with several complementary methods
01 · The analytical chain
HRV is calculated from intervals — but the intervals come from the signal.
A useful way to think about ECG and HRV analysis is as a sequence. Errors introduced near the beginning of the chain can propagate into every metric calculated later.
Detect beats
Identify R peaks consistently and inspect questionable detections.
Build intervals
Calculate the time between successive detected beats.
Quality-control RR
Handle artefacts and inappropriate beats before normal-to-normal analysis.
Interpret HRV
Use several metrics together and compare recordings made under suitable conditions.
Key principle: a sophisticated HRV calculation cannot repair a poor interval series automatically. Signal review and beat-detection quality are part of the analysis, not merely a technical preparation step.
02 · RR and NN intervals
RR is a measurement. NN is an analytical decision.
An RR interval is the time between consecutive detected R peaks. For many HRV calculations, analysts use a normal-to-normal (NN) series after beats or intervals considered unsuitable for the intended analysis have been identified and handled.
Observed beat-to-beat timing
The RR series reflects what the beat detector measured. It can contain true physiological variation as well as detection errors, movement artefacts or intervals influenced by ectopic beats.
Intervals selected for HRV analysis
NN analysis aims to characterize variability between appropriate normal beats. The exact filtering or editing approach should be documented because it can affect the result.
Why visual context matters
A suspiciously long or short interval may represent physiology, an ectopic beat, a missed R peak or a false detection. Looking back at the ECG can help distinguish these possibilities.
03 · Three families of HRV metrics
No single HRV number describes the whole interval series.
Time-domain, frequency-domain and nonlinear methods summarize different properties of beat-to-beat variability. They are complementary rather than interchangeable.
How much the intervals vary
Time-domain measures summarize the magnitude of variability across the analysed interval series.
- SDNN describes dispersion of NN intervals over the analysed period.
- RMSSD emphasizes short-term changes between successive NN intervals.
- Results depend on recording duration and the analysed population.
How variability is distributed across frequencies
Spectral analysis examines oscillatory components in the interval series and how signal power is distributed across frequency bands.
- Respiration can strongly influence higher-frequency variability.
- Recording duration and stationarity affect spectral estimates.
- Individual spectral ratios should not be treated as direct stand-alone measurements of “sympathetic versus parasympathetic balance.”
Pattern, geometry and complexity
Nonlinear methods describe characteristics that may be less visible in simple averages or standard deviations.
- Poincaré plots visualize the relationship between successive intervals.
- SD1 and SD2 characterize different geometric aspects of that distribution.
- Entropy or complexity measures require careful attention to data length and method settings.
04 · ECG morphology and HRV
Waveform shape and beat timing answer different questions.
ECG morphology describes the shape and timing of electrical events within a cardiac cycle. HRV describes variation in timing from beat to beat. Looking at both can provide context that either view alone may miss.
What morphology can add
The ECG trace can help verify whether detected beats correspond to actual cardiac complexes and whether a period of unusual interval variability coincides with signal artefact or altered beat morphology.
- quality of R-peak detection;
- QRS timing and representative beat structure;
- context for irregular or unexpected intervals;
- recognition of segments that may require exclusion from HRV analysis.
What morphology cannot do alone
A single-channel wearable ECG does not provide the spatial information of a standard diagnostic 12-lead ECG. Morphological measurements must therefore be interpreted within the limitations of the acquisition configuration.
- lead placement changes waveform appearance;
- movement and muscle activity can distort morphology;
- software landmarks depend on signal quality;
- wearable recordings do not replace clinical diagnostic assessment.
05 · Repeatable measurements
For trends, consistency of measurement conditions matters.
HRV is sensitive to physiological state and to how the recording is performed. Longitudinal comparison becomes more meaningful when repeated measurements are collected under reasonably similar conditions.
Standardize the context
- use a similar body position;
- allow a comparable settling period;
- record at a similar time of day when possible;
- note recent exercise, stimulants, sleep and unusual stressors.
Protect signal quality
- minimize unnecessary movement;
- maintain reliable electrode contact;
- avoid comparing recordings with very different artefact burdens;
- consider breathing pattern when interpreting respiratory-linked HRV.
Compare like with like
- use comparable analysis duration;
- apply consistent artefact-handling rules;
- compare the same metrics and units;
- interpret trends alongside the original signal and recording context.
Recording length matters. Values derived from ultra-short, short-term and long-term recordings should not be assumed to be directly interchangeable. The measurement protocol belongs to the result.
06 · Interpreting change
A trend is usually more informative than an isolated number.
HRV varies substantially between individuals. A measurement becomes more useful when it is interpreted relative to the person, the recording conditions, the duration of analysis and previous comparable measurements.
Higher is not automatically betterAbnormal rhythms, ectopic activity or detection errors can increase apparent variability. Context and signal quality remain essential.
One threshold rarely fits everyoneAge, physiology, recording duration and methodological choices affect expected values.
Within-person baselines are usefulRepeated standardized recordings can reveal whether a value is typical or unusual for the same individual.
Look at several dimensionsHeart rate, time-domain HRV, spectral structure, nonlinear views and ECG context can complement one another.
07 · Common analytical mistakes
What can make an HRV result misleading.
Trusting every detected beat
Automatic beat detection is useful but not infallible. False or missed detections can create artificial interval variability.
Ignoring recording conditions
Comparing standing after exercise with quiet seated rest may reflect protocol differences more than longitudinal change.
Reducing HRV to one score
A single metric can hide information contained in the interval distribution, spectrum, nonlinear pattern and ECG context.
Using universal cut-offs without context
Reference values depend on population, age, measurement duration and methodology.
Mixing recording durations
Metrics obtained from different observation windows may not represent the same physiological information.
Ignoring the original ECG
When interval behaviour looks unusual, the waveform can help determine whether the cause is physiological or technical.
08 · Further reading
Scientific background.
The following reviews provide useful background on HRV metrics, measurement conditions and the importance of data quality.
Overview of time-domain, frequency-domain and nonlinear HRV metrics, recording durations and normative considerations.
Recommendations for planning, measurement, analysis and reporting in HRV research.
Discussion of HRV methodology, measurement periods and the effect of R-peak recognition errors on analysis.
09 · Frequently asked questions
ECG, RR intervals and HRV.
What is the difference between RR intervals and NN intervals?
RR intervals describe the timing between detected R peaks. NN intervals are the intervals retained for normal-to-normal HRV analysis after unsuitable beats or artefacts have been handled according to the chosen analysis method.
Is higher HRV always better?
No. HRV is context-dependent. Signal artefacts, ectopic beats and rhythm disturbances can increase apparent variability, and expected values depend on factors such as recording duration, age and physiological state.
Can a one-minute HRV value be compared directly with a five-minute value?
Not automatically. Measurement duration affects HRV metrics, and reference values from different recording lengths are not necessarily interchangeable.
Why review ECG if the goal is HRV?
Because HRV depends on the interval series. Reviewing the ECG can help identify missed beats, false R-peak detections, artefacts or unusual beat morphology that may distort derived variability metrics.
Continue from the science to the KardioAS workflow.
Explore how the platform separates signal review, ECG morphology, HRV, comparison, trends and safety limitations.