How does the heart monitor work when your watch shows a sudden jump the moment you stand up? The number looks direct and authoritative, but most wrist devices don't listen to your heart's electrical activity at all. They shine light into your skin, watch how blood volume changes, and use software to infer a pulse rate.
That distinction matters. A reading can be useful without being a direct measurement, and it can also become misleading when movement, poor circulation, loose contact, or an unusual posture changes the signal. Understanding the difference between optical pulse sensing and electrical heart sensing helps you know when a wearable reading is a helpful trend, when it needs context, and when it deserves clinical follow-up.
What Your Watch Is Actually Sensing During a Heart Rate Spike
Your watch buzzes shortly after you stand. You feel lightheaded, your heart seems to race, and the screen shows a sharp increase in beats per minute. It's natural to think the watch has watched your heart speed up directly.
It hasn't.
Inside the back of the watch, a light source shines into the skin while a photodetector watches the returning light. Arterial blood volume changes slightly with each pulse, altering how much light returns to the sensor. The device turns that changing optical pattern into an estimated pulse rate. This explanation of wrist-based heart-rate tracking shows why the displayed number is a computed result rather than a direct view of the heart.
The moment behind the number
The process happens quickly:
- Light enters the wrist. The sensor illuminates nearby tissue.
- Blood changes the optical signal. Each arterial pulse changes local blood volume and the amount of reflected light.
- The detector captures variation. A photodetector converts returning light into an electrical signal.
- Software identifies timing. Algorithms estimate the interval between pulses and convert it into beats per minute.
- The interface displays a result. An alert may appear when the calculated rate crosses a programmed threshold.
So the watch is measuring a peripheral pulse signal. It's using that signal as a practical proxy for heart rate because each effective heartbeat usually creates a corresponding arterial pulse.
The practical distinction: Your watch usually sees the consequence of the heartbeat in the wrist, not the electrical event that initiated it.
That difference becomes important during a POTS or dysautonomia episode. A rapid rise after standing may be a meaningful pattern, but the watch still sees it through changing blood flow at the wrist. Posture, skin temperature, contact pressure, and movement can all affect what the sensor receives. A timeline therefore shows a derived signal that needs interpretation, not a perfect recording of every cardiac event.
The Optical Path Inside a PPG Sensor
Think of a PPG sensor as a flashlight and a light meter working together. Shine a flashlight through your finger, and the amount of light reaching the other side changes as blood pulses through the tissue. A wrist device uses the same broad idea, although it generally measures light reflected back toward the sensor rather than relying only on transmitted light.
PPG, or photoplethysmography, follows a chain from illumination to calculation:
From LED to returning photons
The LED sends light into the skin. Some photons scatter through the epidermis, dermis, and small blood vessels, while others are absorbed by tissue, venous blood, or pulsating arterial blood. The photodetector captures the photons that return, then converts their intensity into a changing voltage signal.
Many wearable designs use several wavelengths for different sensing tasks. Green light is commonly useful for shallow blood-flow changes near the skin, while red and infrared light can provide information from deeper optical paths and support other measurements. The exact arrangement varies by device, so a consumer watch shouldn't be assumed to use every wavelength in the same way.
Separating the pulse from the background
The raw optical signal contains two useful layers:
- The DC component is the larger, relatively steady baseline created by static tissue, venous blood, and overall light absorption.
- The AC component is the smaller pulsatile change associated with arterial blood-volume variation.
The algorithm looks for repeating changes in the AC component. It estimates the timing between peaks, smooths irregular points, rejects obvious artifacts, and translates the intervals into a heart-rate estimate. This is why a stable fit and good skin contact matter. The device needs to distinguish a small pulse-related modulation from a much larger background signal.
Green light often dominates wrist heart-rate sensing because it can produce strong contrast from shallow perfusion near the skin. That convenience comes with a tradeoff. PPG remains indirect, so local circulation and optical coupling can affect the result more than they would with electrodes that measure electrical activity directly.
A pulse-ox explanation can make the broader optical-sensing idea easier to follow, especially when you want to understand how reflected light carries physiological information. This guide to the PI signal in pulse oximetry provides useful context for that relationship.
ECG Leads and the Electrical View of the Same Heart
PPG watches blood-volume change. An electrocardiogram, or ECG, records the electrical activity that causes the heart muscle to contract. The two signals belong to the same heartbeat, but they answer different questions.
A wearable ECG generally uses a circuit completed through the user's body. One electrode contacts the skin through the device, and another contact is touched with a finger from the opposite hand. That arrangement captures an electrical difference across the body that resembles a single-lead view. It can produce a waveform rather than only a pulse estimate.
What an electrical trace can reveal
An ECG waveform may show features such as the P wave, the QRS complex, the PR interval, and the QT duration. Those features help a clinician assess whether atrial activation appears present, how ventricular depolarization looks, and whether the timing between electrical events is unusual.
PPG doesn't provide that same waveform detail. It can show that pulses are arriving at particular intervals, but it can't directly show whether a P wave preceded each ventricular beat or whether the QRS shape has changed. That limitation is central when distinguishing a wellness trend from rhythm evidence. A resource on atrial and ventricular rate differences illustrates why electrical timing and peripheral pulse timing shouldn't be treated as interchangeable.
| Attribute | PPG (Optical) | Single-Lead ECG (Electrical) |
|---|---|---|
| Primary signal | Arterial blood-volume change | Electrical voltage difference |
| What it estimates or records | Pulse timing and estimated heart rate | Cardiac electrical waveform |
| Main strength | Passive, low-power, continuous tracking | More direct rhythm assessment |
| What may disrupt it | Motion, contact, perfusion, optical interference | Poor electrode contact and electrical noise |
| Rhythm detail | Timing patterns without full waveform morphology | More information about wave and interval structure |
| Clinical boundary | Useful for trends and alerts, not a complete rhythm diagnosis | Useful for confirming certain rhythm patterns, but not equivalent to a hospital ECG |
A single-lead ECG still isn't a full spatial assessment of the heart. A hospital ECG uses multiple leads to view electrical activity from different directions, which matters for questions such as ischemia or conduction abnormalities. The wrist ECG can be valuable, but it shouldn't be mistaken for a complete cardiac evaluation.
Where Wrist Monitoring Breaks Down
A watch can show a precise-looking number while the underlying pulse signal is unstable. Wrist PPG works best when the sensor stays still against well-perfused skin. As movement increases, arm motion, tissue deformation, and changing contact can reshape the optical waveform. One review reported mean absolute percentage errors above 20% during vigorous activity in its discussion of wearable heart-rate accuracy.
The common failure modes
Motion artifact is the most familiar problem. Walking, typing, lifting, or running can create optical changes that resemble extra beats, or hide genuine peaks. During vigorous exercise, a high displayed rate may therefore reflect movement-related distortion rather than a matching electrical rhythm.
Low perfusion weakens the signal in a different way. Cold hands, vasoconstriction, or a vasovagal episode can reduce the pulsatile blood reaching the detector. The pulsatile AC component then becomes small beside the DC background, leaving the algorithm with less dependable information.
Fit and optical coupling also shape the result. A loose band can move separately from the skin, while excessive pressure can alter the tissue beneath the sensor. Tattoos, hair, sweat, and uneven contact may interfere with the light path, although the effect differs by device and skin.
Irregular beats create a timing problem between the heart and the wrist. A premature beat may produce a weak peripheral pulse, followed by a compensatory pause and a long interval. The electrical event can occur even when the corresponding pulse is faint or missing at the wrist.
| Failure Mode | Typical Error Range | Mitigation |
|---|---|---|
| Vigorous movement | Mean absolute percentage error can exceed 20% in reported review findings | Treat exercise readings cautiously, keep the device stable, and review trends rather than isolated points |
| Walking, typing, or arm motion | Accuracy generally declines as activity increases | Pause for a still reading, use a secure fit, and check whether the value matches how you feel |
| Cold or poorly perfused skin | No single universal error range applies | Warm the hand, wait for better circulation, and avoid interpreting a weak-signal reading alone |
| Loose or shifting band | Device-specific and signal-dependent | Move the watch slightly above the wrist bone and secure it without excessive pressure |
| Irregular pulse transmission | Varies with the beat pattern | Save an ECG tracing when available and discuss recurring symptoms with a clinician |
A sensible algorithm may suppress a reading instead of displaying a confident-looking number. A missing value can be more honest than a precise-looking artifact. For symptom-aware monitoring, that uncertainty matters: a questionable pulse should be marked as questionable, so a later episode summary is not built from noise.
From Beats to Episodes The Detection Algorithm
A heart-rate stream becomes clinically more useful when software groups individual readings into an event with a beginning, a rise, a duration, and a context. The basic pipeline starts with pulse peaks, converts them into inter-beat intervals, and then asks whether the current pattern differs meaningfully from the person's recent baseline.
Use a 30-plus beats-per-minute rise within five minutes as a worked example. This criterion is part of Cardiogram's stated detection approach for identifying tachycardic episodes, but the logic is easier to understand as a sequence rather than as a single alert.
Following the signal through the pipeline
- Beat detection marks peaks. The algorithm identifies likely pulse peaks in the PPG waveform. With ECG, the corresponding timing target would be the electrical R wave. With PPG, it is an optical pulse peak that may arrive slightly later and may be weakened by circulation or motion.
- Inter-beat intervals are calculated. The system measures the time between successive detected beats. Shorter intervals generally indicate a faster estimated rate.
- A baseline is established. Recent resting or low-motion data provides a comparison point. A current reading is more informative when the system knows what is typical for that person at that time.
- The moving window checks the rise. The algorithm compares the running rate with the baseline across the relevant time window. A sustained increase that meets the threshold can trigger an episode flag.
- The event remains active while the pattern persists. The system records timing and duration, then ends the episode after the signal returns toward baseline or the required cool-down condition is met.
Why one threshold isn't enough
A useful system adds checks before labeling an event. Motion-state gating can identify periods when the wearer is moving heavily. Step data can provide another clue, and minimum-duration filters can prevent one noisy point from becoming a full episode. Workout and recovery periods may also need separate handling because exercise naturally raises pulse rate.
Context changes the interpretation. A rise while climbing stairs isn't the same kind of event as a rise after standing. Anxiety, pain, dehydration, medication changes, sleep loss, and orthostatic symptoms can all sit behind a similar numerical pattern. A symptom-aware app can let the user mark what was happening so the episode record reflects more than the pulse alone.
Turning Watch Data Into Clinician-Ready Context
Raw heart-rate history becomes more useful when it answers questions a clinician can act on. Instead of bringing a vague memory of feeling unwell, a patient can connect an episode with posture, symptoms, hydration, medication, sleep, or activity.
The workflow starts with automatic collection, but it shouldn't end there. Manual context gives the graph meaning:
- Posture and timing: Record whether symptoms began after standing, walking, sitting, or lying down.
- Symptoms: Mark dizziness, palpitations, fatigue, brain fog, or other sensations near the event.
- Possible triggers: Note hydration, salt intake, sleep, medication timing, heat, or illness.
- Activity status: Separate planned exercise from an unexpected rise during ordinary daily movement.
Filtering before counting
Exercise can create a legitimate pulse rise that resembles tachycardia in a graph. A context-aware system can exclude workouts and recovery periods from episode counts, reducing the chance that training data overwhelms the pattern a patient and clinician are trying to evaluate.
That matters for POTS and dysautonomia, where the question often isn't, “How high did the rate go?” The more useful questions include: Did the rise follow a position change? How long did it last? Did symptoms occur at the same time? Did it recur under similar conditions?
Building a record someone else can read
Daily and weekly summaries can reveal timing patterns that disappear inside a long stream of individual readings. Episode feeds, resting-rate trends, and heatmaps can help organize the history, while an exportable report can place episodes, symptoms, and trend charts in one document for clinical review.
Cardiogram reads heart-rate data from Apple Health, detects 30-plus beats-per-minute rises within five minutes, records baseline, peak, duration, and occurrence time, and supports symptom and trigger logging. Its stated workflow also includes workout and recovery exclusion, trend summaries, and PDF reporting, with analysis performed on the device and synchronization through the user's iCloud account.
The report doesn't diagnose POTS or replace a medical assessment. It can, however, replace “I felt weird on Tuesday” with a timestamped pattern that gives the appointment a more precise starting point.
What the Numbers Can and Cannot Tell You
A wearable can be useful without being a diagnostic instrument. That may sound like a contradiction, but it's the central idea behind responsible heart monitoring. A pulse trend can reveal that something changed, while the device may still be unable to explain why.
PPG-derived data can show repeated rate patterns, periods of sustained elevation, and possible irregularity. A wearable ECG may provide a more direct rhythm tracing and can support evaluation of certain patterns. Neither result should be treated as a final verdict without clinical correlation.
A reading is evidence, not an answer
| Claim | What the Watch Can Show | What It Cannot Confirm |
|---|---|---|
| “My heart rate rose after standing” | A pulse-rate change associated with a time on the timeline | The cause of the rise or whether the change meets a formal diagnostic standard |
| “My rhythm may be irregular” | An irregular pulse pattern or an alert generated from that pattern | Every arrhythmia, the full electrical mechanism, or the clinical significance |
| “I had sustained tachycardia” | A prolonged elevated pulse estimate when signal quality is adequate | Whether the rhythm is sinus tachycardia, another rhythm, or a response to an underlying condition |
| “The ECG looks abnormal” | A single-lead electrical tracing with visible waveform features | What a full multi-lead assessment, imaging, laboratory testing, or specialist review would show |
| “I don't have a heart problem because the watch was normal” | A normal reading during the period the device captured | Silent events, intermittent arrhythmias, structural disease, or ischemia outside the recording window |
The history of heart monitoring helps explain why no single wearable view answers every question. The first human ECG recordings appeared in the late nineteenth century, and later work on the string galvanometer established the clinical principle of detecting tiny voltage differences at the body surface. A historical review of ECG development describes that progression. Portable continuous monitoring arrived much later, with Norman Holter's dynamic ECG invented in 1957 and early equipment weighing more than 38 kilograms, as detailed in this history of ambulatory monitoring.
That history reinforces a practical point. Monitoring is valuable because it captures events that a brief appointment might miss, but the capture method still determines what the result means. Wrist PPG is especially helpful for trends and episode timing, while electrical recordings provide a different layer of rhythm information.
If you receive an alert or see a troubling pattern, save the relevant trace, write down what you felt and what you were doing, and share the record with a clinician. They can decide whether you need further evaluation, such as laboratory testing, longer rhythm monitoring, or an assessment designed for orthostatic symptoms. Seek urgent medical care for severe or rapidly worsening symptoms, especially fainting, chest pain, serious breathing difficulty, or signs of a medical emergency.
Cardiogram turns Apple Watch heart-rate history into structured episodes, symptom-linked context, and clinician-ready summaries for people investigating POTS, dysautonomia, or unexplained tachycardia. Visit Cardiogram to see how its episode detection and reporting workflow can help you organize the data before your next clinical conversation.
