Have you ever wondered what your heart could tell you about your overall health and vitality? Heart rate variability (HRV) offers a unique window into the intricate balance of your body’s autonomic nervous system, reflecting how well you adapt to life’s daily demands. HRV is more than just a measure of your heartbeats—it's a key indicator of your resilience, recovery, and overall well-being.
In this article, we’ll dive into the fascinating world of HRV, exploring its fundamental concepts, important measurements and have a look at frequency bands, their calculation and visualisation.
Heart rate variability (HRV) is the variation in time between consecutive heartbeats. While your heart might beat at an average rate of, say, 60 beats per minute, the actual time between beats isn’t perfectly consistent—it might fluctuate slightly, like 0.98 seconds between one pair of beats and 1.02 seconds between the next. This natural variation is what we call HRV, and it offers valuable insights into your body’s health and resilience. A near-zero HRV often means that your sympathetic nervous system (fight-or-flight mode) is completely dominant, leaving your body in a state of high stress with no capacity for recovery. It can occur in cases of severe burnout, extreme overtraining, or chronic emotional stress.
At its core, HRV reflects the balance of your autonomic nervous system (ANS), which has two main branches:
- Sympathetic Nervous System (SNS): Often called the "fight or flight" system, it’s responsible for energising your body during stress or activity.
- Parasympathetic Nervous System (PNS): Known as the "rest and digest" system, it promotes relaxation, recovery, and repair.
A healthy balance between these two systems allows your body to adapt effectively to life’s challenges—whether it’s responding to a stressful situation or recovering after exercise. HRV acts as a mirror of this balance: a higher HRV generally indicates greater adaptability and resilience, while a lower HRV might signal stress, fatigue, or health issues.
Understanding these metrics can help you tune into your body’s needs. For example:
- Feeling stressed? A low HRV might signal the need for relaxation techniques like deep breathing or mindfulness.
- Planning a workout? A higher RMSSD (Root Mean Square of Successive Differences) may suggest your body is ready to perform, while a lower score might mean you need more rest.
HRV isn’t just for athletes or data enthusiasts—it’s a practical tool for anyone seeking to improve their well-being. By learning to listen to your heart’s rhythm, you can make smarter decisions about your health, stress, and energy levels.
Measuring heart rate variability (HRV) has become more accessible than ever, thanks to advancements in wearable technology and mobile apps: Wearables for HRV tracking include smartwatches from brands like Garmin, Fitbit, and Apple, as well as chest straps like the Polar H10, which are highly accurate and favoured by athletes. Additionally, mobile apps such as Elite HRV, and HRV4Training, along with dedicated devices like the Oura Ring, provide detailed HRV analysis and insights into sleep and recovery.
To get reliable HRV data, consistency and proper technique are crucial:
Consistency is Key:
Avoid Confounding Factors:
Use Long-Term Trends:
You can also evaluate HRV on your own using Python - devices like the H10 chest strap allow to extract a comma separated value (csv) file listing purely the durations between two consecutive peaks during your measurement window. From this minimal set of information you can extract a lot of information - let's dive into this:
How does the input data look like?
Based on this you can easily calculate a set of time-domain metrics which already provide valuable insights into the activity of your autonomic nervous system:
def calculate_metrics(rr):
"""Calculate time-domain HRV metrics."""
results = {
'Mean RR (ms)': np.mean(rr),
'STD RR/SDNN (ms)': np.std(rr),
'Mean HR (beats/min)': 60000 / np.mean(rr),
'STD HR (beats/min)': np.std(60000 / rr),
'RMSSD (ms)': np.sqrt(np.mean(np.square(np.diff(rr)))),
}
return results
STD RR / SDNN (ms)
A time-domain measure of overall heart rate variability measuring the standard deviation of the time between normal heartbeats (NN intervals) over a specified period. It captures both short-term and long-term fluctuations. Higher SDNN indicates greater HRV and better adaptability of the autonomic nervous system, while lower SDNN may indicate stress, fatigue, or poor autonomic regulation.
RMSSD (ms)
A short-term HRV metric that reflects parasympathetic (rest-and-digest) activity of the autonomic nervous system. A higher RMSSD indicates better recovery and parasympathetic activity. Often associated with relaxation or good physical conditioning, while a lower may suggest stress, fatigue, or insufficient recovery.
Time domain metrics provide valuable information about overall heart rate variability, but to understand the underlying physiological mechanisms more deeply, we need to examine the frequency components of HRV signals.
Frequency domain analysis reveals physiological insights that time-domain metrics cannot capture by decomposing HRV into specific frequency bands, enabling more precise assessment of autonomic regulation, stress levels, recovery capacity, and long-term health outcomes.
The key frequency bands in HRV analysis are:Frequency domain analysis is a valuable tool for monitoring stress, recovery, and autonomic balance: A shift toward lower high-frequency (HF) power or an elevated low-frequency to high-frequency (LF/HF) ratio may indicate increased stress or sympathetic dominance, while higher HF power suggests parasympathetic activation and readiness for rest and recovery. In sports performance, increased HF power signals readiness for intense activity, whereas elevated LF power or reduced HF power highlights the need for rest. By combining frequency domain metrics with time-domain metrics, athletes and individuals gain a comprehensive understanding of autonomic regulation, optimising training loads and recovery while avoiding overtraining.
The notebook Exploring Heart Rate Variability (HRV) provides very helpful insights into domain frequency analysis and how it can be calculated. Building on this foundational knowledge, I'll now apply these concepts to analyse a longer-duration HRV recording of approximately 7 hours, creating frequency domain visualisations through heatmaps. Here's the methodological approach I implemented:
Remarks on the preprocessing step: Filtering eliminates random fluctuations caused by measurement errors or transient physiological events. You could for instance apply a moving average to the data using below code which slides the weights array over the input data:
python def moving_average(data, window_size): return np.convolve(data, np.ones(window_size) / window_size, mode='valid')
Another preprocessing step is called 'detrending' which removes the non-stationary trends from the RR series since trends may cause distortions to time-and-frequency domain analysis. You can achieve this by applying a Savitzky-Golay filter removing the trend from the original data. Below you see a picture showing the data series (RR) before (black) and after detrending (green):
Once all the preprocessing steps have been performed you can apply the Fourier Transformation to the dataset to obtain the power spectral density for each window. The last step is then the visualisation of the power:
X-axis (time in minutes): Tracks the progression of HRV metrics over time, giving a continuous representation of changes in autonomic nervous system activity.
Y-axis (frequency ranges): Represents different frequency bands of HRV
High PSD in VLF Range (Blue Intensity):
Dynamic LF Activity (Dark Red to Yellow):
Lower HF Power (Yellow to Grey/White):
Following is the code implementation that generated the heatmap visualization shown above:
def generate_heatmap(result_freq, fs, elements, hex_list):
# generates a continuous color map based on these colors
color_map = get_continuous_cmap(hex_list)
color_axes = np.linspace(0, 1, len(hex_list))
# The color spectra values correspond to specific frequencies in the data
color_spectra = [0, 10, 30, 60, 180, 240, 320, 400, 1200]
# Maps the color_spectra values to the color_axes values linearly
f1 = interp1d(color_spectra, color_axes, kind='linear', bounds_error=False, fill_value=1)
d_points_constant_welch = 129
color_array = np.empty([d_points_constant_welch, 0]) # Predefine the array
for i in range(elements):
pxx = result_freq[i]['pxx']
# Intrapolates power density to color map value range 0..1
scaled_colors = f1(pxx)
color_array = np.append(color_array, np.expand_dims(scaled_colors, axis=1), axis=1)
# Optional: Apply Gaussian filter - reduce noise and detail, making the data smoother
data = gaussian_filter(color_array, sigma=0.8)
# Adjust x and y dimensions to align with data
x = np.linspace(0, elements * 5, data.shape[1] + 1) # One more than columns
y = np.linspace(0, 2, data.shape[0] + 1) # One more than rows
fig, ax = plt.subplots()
quad_mesh = ax.pcolormesh(x, y, data, cmap=color_map, shading='flat') # Use 'flat' shading
color_bar = fig.colorbar(quad_mesh, ticks=color_axes)
color_bar.ax.set_yticklabels(color_spectra)
ax.set_xlim(0, elements * 5)
ax.set_ylim(0, 2)
plt.title("Heatmap of Frequency-Domain Metrics")
plt.show()Heart rate variability (HRV) is more than a metric—it’s a window into your body’s inner workings and a guide to living a healthier, more balanced life. By tracking your HRV daily, linking it to your lifestyle choices, and adjusting habits like sleep, exercise, and stress management based on your data, you can take control of your health and well-being in ways you never thought possible.
Developing your own “HRV routine” is simpler than it sounds. Start by measuring your HRV consistently, observe trends over time, and use these insights to make informed decisions. Whether it’s adjusting your workouts to prevent burnout or recognising when your body needs rest, HRV empowers you to align your actions with your body’s needs.
The benefits are clear: better health, enhanced productivity, reduced stress, and improved performance in both daily life and sports. HRV provides a roadmap for achieving emotional resilience, physical vitality, and overall quality of life.
So why wait? Start tracking your HRV today to unlock the full potential of your health and performance.
Alberto Desiderio is deeply passionate about data analytics, particularly in the contexts of financial investment, sports, and geospatial data. He thrives on projects that blend these domains, uncovering insights that drive smarter financial decisions, optimise athletic performance, or reveal geographic trends.