Atrial fibrillation shows up on a pulse oximeter as an erratic, uneven pulse waveform and a heart rate reading that jumps around from beat to beat, rather than holding steady. The little wave tracing on the screen, if your device displays one, loses its regular, repeating pattern and instead shows peaks that vary in height and spacing. A standard fingertip pulse oximeter was never designed to diagnose AFib, but the visual clues it provides are surprisingly telling once you know what to look for.
How a Pulse Oximeter Picks Up Your Heartbeat
A pulse oximeter works by shining light through your fingertip (or earlobe, or toe) and measuring how much of that light is absorbed by your blood. The core technology is called photoplethysmography, or PPG, and it detects tiny changes in blood volume with each heartbeat.1PubMed Central. Photoplethysmography for blood volumes and oxygenation changes during intermittent vascular occlusions Every time the heart contracts and pushes a pulse of blood into the arteries, the volume of blood in your fingertip increases slightly, absorbing a bit more light. Between beats, it decreases. The oximeter translates these fluctuations into two things: an oxygen saturation number and a pulse rate.
The waveform you see on the screen, sometimes called a plethysmograph or “pleth” wave, is essentially a real-time graph of those blood volume changes. In a healthy, regular rhythm, it looks like a series of smooth, evenly spaced humps, almost like gentle rolling hills. Each peak lines up with a heartbeat, and the valleys between peaks are roughly equal in depth and timing. The pulsing component of this waveform is driven by the heart’s beat-to-beat output, while a slower baseline shift reflects things like breathing and changes in blood vessel tone.2Physiological Measurement. Photoplethysmography and its application in clinical physiological measurement
What AFib Does to the Pulse Signal
In atrial fibrillation, the heart’s upper chambers quiver chaotically instead of contracting in an organized way. This means the ventricles, the main pumping chambers, receive irregular signals about when to contract. The result is a heartbeat that is “irregularly irregular,” a phrase doctors use because there is no predictable pattern to the timing at all. Some beats come quickly after the previous one, others arrive after a longer pause.
This irregularity has a direct physical consequence for what the pulse oximeter detects. When two beats are close together, the heart has had less time to fill with blood, so the second beat pumps out a smaller volume. When a longer gap occurs, the heart fills more completely and ejects a bigger volume. Research using Doppler ultrasound in people with chronic AFib has shown that this beat-to-beat variation in the amount of blood pumped out actually increases at higher heart rates, even though the spacing between beats becomes less variable.3PubMed. Influence of heart rate on stroke volume variability in atrial fibrillation in patients with normal and impaired left ventricular function In practical terms, a faster AFib episode produces more dramatic swings in pulse strength from one beat to the next.
This is why AFib does not just look like a slightly off rhythm on a pulse oximeter. It disrupts both the timing and the amplitude of the waveform, creating a signal that is visibly messy compared to a normal sinus rhythm.
What You Actually See on the Screen
If your pulse oximeter has a waveform display, the most obvious sign of AFib is that the peaks are no longer evenly spaced. Instead of that neat row of matching hills, you get peaks arriving at random-looking intervals. Some will be tall, some noticeably shorter, and the gaps between them will vary. The waveform looks jittery and unsettled compared to the metronomic pattern of a normal heartbeat.
The heart rate number itself becomes unreliable in a different way than you might expect. Most consumer pulse oximeters calculate heart rate by averaging over several seconds. During AFib, the instantaneous rate can swing wildly, say from 60 to 130 within a short window, but the displayed number may bounce around or settle on a value that does not match what you feel at your wrist. If your oximeter updates the rate frequently, you might notice it flickering between very different numbers every few seconds.
Oxygen saturation readings can also behave oddly during AFib, though not always. When beats are very irregular, the oximeter’s algorithm may struggle to lock onto a clean signal, and you might see the SpO2 number dip briefly or display dashes. This does not necessarily mean your oxygen is actually low. It often reflects the device having trouble processing a chaotic pulse rather than a true drop in blood oxygen. That said, AFib can genuinely reduce how efficiently the heart pumps blood in some people, so persistently low readings during an episode should not be dismissed.
Can a Standard Fingertip Pulse Oximeter Detect AFib?
A typical $20 fingertip pulse oximeter from the pharmacy was built to measure oxygen saturation, not to diagnose heart rhythms. It does not have software that analyzes rhythm regularity or flags arrhythmias. So while AFib produces visible abnormalities in the waveform and pulse rate display, the device itself will not tell you “atrial fibrillation detected.” You are left interpreting the raw visual output on your own, which is unreliable for obvious reasons.
That said, a newer category of AI-enhanced pulse oximeters is changing this picture. One real-world study tested a pulse oximeter equipped with an artificial intelligence algorithm specifically designed to flag AFib. When compared against expert-interpreted electrocardiograms, the device achieved about 97% sensitivity and 91% specificity, meaning it caught nearly all true AFib cases while incorrectly flagging only about one in ten people who did not have it.4PubMed. Diagnostic accuracy of an artificial intelligence-enhanced pulse oximeter for atrial fibrillation detection: a real-world population study The device maintained strong accuracy even across different clinical subgroups, though its performance dipped slightly in people with very low perfusion, meaning poor blood flow to the fingertip.
The distinction matters. If you are using a basic consumer oximeter, you are looking for visual clues and should treat anything suspicious as a reason to seek a proper evaluation, not as a diagnosis. An AI-enhanced oximeter with a validated AFib algorithm is a fundamentally different tool, closer to a screening device than a simple oxygen monitor.
How Smartwatch Detection Compares
Many people first encounter AFib screening not through a pulse oximeter but through a smartwatch. Smartwatches use the same underlying PPG technology, shining green light against the skin of the wrist and measuring blood volume changes. The hardware is conceptually identical to what sits inside a fingertip pulse oximeter, though the signal quality differs because the wrist is a noisier measurement site with more motion artifact than a fingertip.
The algorithms running on smartwatches, however, are purpose-built for rhythm analysis. They extract features from the pulse signal such as how much the intervals between beats vary and how random the pattern is. Higher variability and greater randomness point toward AFib, since a normal rhythm is relatively predictable while AFib is not.5Scientific Reports. Atrial Fibrillation Detection from Wrist Photoplethysmography Signals Using Smartwatches Research on fingertip PPG signals has shown that analyzing just two minutes of data can reliably separate AFib from normal rhythm, with the analysis performing nearly as well as using ten minutes of data.6Scientific Reports. Identification of Atrial Fibrillation by Quantitative Analyses of Fingertip Photoplethysmogram
A systematic review and meta-analysis comparing PPG-based smartwatches to ECG-based smartwatches found that PPG devices had pooled sensitivity around 97% and specificity around 97%, outperforming the ECG-equipped watches, which came in at roughly 83% sensitivity and 88% specificity.7PubMed Central. Comparison of diagnostic accuracy of electrocardiogram-based versus photoplethysmography-based smartwatches for atrial fibrillation detection: A Systematic Review and Meta-Analysis That result is somewhat counterintuitive since ECGs are the gold standard for diagnosing arrhythmias in clinical settings. The likely explanation is that PPG-based monitoring runs continuously in the background, sampling thousands of moments, while smartwatch ECGs require the user to actively touch the watch and hold still for 30 seconds. Continuous passive monitoring catches more episodes, including brief ones the user never notices.
Ambulatory validation of a smartwatch using simultaneous Holter monitor recording as the reference standard found sensitivity of about 97% and specificity around 89% for detecting AFib.8American Heart Journal. Atrial fibrillation detection using ambulatory smartwatch photoplethysmography and validation with simultaneous holter recording These numbers are encouraging but come with a caveat: performance in free-living conditions, where people are moving, exercising, and going about their day, is harder to nail than performance in a controlled clinical setting.
When Irregular Readings Are Not AFib
Not every erratic waveform on a pulse oximeter means AFib. Several common situations can produce a messy signal that might look alarming:
- Motion artifact: Moving your hand, shivering, or even tapping your foot can shake the sensor enough to create irregular-looking peaks. This is the most common cause of a jumpy reading and disappears when you hold still.
- Poor perfusion: If your hands are cold, your blood pressure is low, or you have peripheral vascular disease, the pulse signal reaching your fingertip is weak. The oximeter may struggle to distinguish real beats from noise, producing a waveform that looks chaotic even though your heart rhythm is perfectly regular.
- Premature beats: Occasional premature atrial or ventricular contractions (PACs or PVCs) are extremely common and usually harmless. They show up as an early, often smaller peak followed by a compensatory pause and then a larger beat. This can look intermittently irregular, but the pattern is different from AFib’s sustained randomness. A premature beat creates a recognizable “short-long” pair, while AFib produces chaos with no repeating pattern.
- Nail polish or artificial nails: Dark nail polish, gel nails, or acrylics can interfere with light transmission, causing erratic readings. The waveform quality may degrade without any cardiac cause.
- Respiratory variation: Normal breathing causes slight cyclical changes in heart rate (the rate increases slightly when you inhale and decreases when you exhale). In some people, especially younger, fit individuals, this sinus arrhythmia can be pronounced enough to make the waveform look mildly irregular.
The key difference is that AFib produces sustained irregularity that persists when you are sitting still, warm, and relaxed. If the erratic pattern goes away when you warm your hands and stop moving, it was almost certainly not AFib. If it persists for minutes at a time despite ideal conditions, that warrants further investigation.
Why Perfusion Matters for Accuracy
The strength of the blood flow signal at your fingertip, measured as the perfusion index, affects how well any PPG-based device can detect rhythm abnormalities. A perfusion index below about 1% means the pulsing component of the signal is very small relative to the background, making it harder for the device to accurately distinguish one beat from the next. In studies of AI-enhanced pulse oximeters, diagnostic accuracy remained high across most clinical groups but showed a slight decrease in people with very low perfusion.4PubMed. Diagnostic accuracy of an artificial intelligence-enhanced pulse oximeter for atrial fibrillation detection: a real-world population study
This matters practically because many of the people at highest risk for AFib, older adults with heart failure, peripheral artery disease, or diabetes, are also the people most likely to have poor peripheral perfusion. If you are using a pulse oximeter to monitor your rhythm and your hands tend to be cold or the device often shows a weak signal, the readings deserve extra skepticism. Warming your hands before taking a reading, using a well-fitted sensor, and sitting still for at least a minute can all improve signal quality.
Camera-Based Detection Without Touching Anything
An emerging technology takes the same PPG principle and applies it using an ordinary camera instead of a fingertip sensor. By recording your face on video, subtle color changes in the skin caused by pulsing blood flow can be extracted and analyzed the same way a pulse oximeter analyzes its light signal. This technique, called remote PPG, has been tested for AFib detection. One study using 10 minutes of facial video recording combined with deep learning algorithms achieved about 93% sensitivity and 98% specificity in distinguishing people with AFib from those without it, using 12-lead ECGs as the reference standard.9Scientific Reports. Contactless facial video recording with deep learning models for the detection of atrial fibrillation
This is still a research tool, not something available in consumer products yet, but it hints at a future where a phone’s front-facing camera could screen for AFib during a video call or while you scroll through social media. The accuracy is not quite as high as direct-contact PPG, partly because ambient lighting, facial movement, and skin tone all introduce variability that a clipped-on sensor avoids. Still, the idea of completely passive, contactless arrhythmia screening is a compelling direction, especially for populations who would not wear a smartwatch or remember to clip on a pulse oximeter.
What to Do If Your Pulse Oximeter Looks Off
If you notice your pulse oximeter’s waveform looking persistently irregular or the heart rate display jumping around, here is a reasonable approach. First, rule out the obvious: sit still, warm your hands, remove nail polish from the measurement finger, and try again. If the irregularity persists for several minutes under good conditions, take note of what the display shows and, if possible, what your heart rate numbers are doing.
A pulse oximeter reading, no matter how suspicious, is not a diagnosis. AFib can only be confirmed with an electrocardiogram, which measures the heart’s electrical activity directly rather than inferring it from blood flow in a fingertip. What a pulse oximeter can do is give you a reason to seek that ECG sooner rather than later. Many people with AFib have episodes that come and go, sometimes lasting only minutes. If your oximeter catches an irregular pattern during one of those episodes, mentioning it to your doctor can prompt monitoring that might otherwise not happen.
For people already diagnosed with AFib, a pulse oximeter can serve as a rough, informal check on rate control. If you are on medications to keep your heart rate in a target range, seeing the displayed rate consistently above that range, or seeing a wildly erratic waveform when it had been smoother, gives you something concrete to discuss at your next appointment. It is not a substitute for the monitoring your cardiologist recommends, but it adds a data point you can gather at home without any special equipment.
Paroxysmal AFib and the Detection Gap
One of the trickiest aspects of atrial fibrillation is that many people have the paroxysmal form, meaning it comes and goes unpredictably. An episode might last minutes to hours, then the heart snaps back to a normal rhythm. Between episodes, the pulse oximeter waveform will look completely normal because the heart is, in fact, beating normally. This creates a detection gap: if you happen to check during a normal-rhythm window, you will see nothing abnormal.
This is where continuous or frequent monitoring has an edge over spot checks. A single pulse oximeter reading is a snapshot. If AFib is intermittent, the snapshot may miss it entirely. Smartwatches that sample the pulse throughout the day have a much better chance of catching a brief episode because they are always watching. The large prospective trials run by major smartwatch manufacturers enrolled hundreds of thousands of participants and flagged many previously unknown AFib cases, precisely because passive monitoring catches what a doctor’s office visit or a single home reading cannot.
If you suspect you might have intermittent AFib based on occasional symptoms like a racing or fluttering heartbeat, dizziness, or unexplained fatigue, taking a pulse oximeter reading during the symptoms is far more informative than taking one when you feel fine. The same goes for any wearable. An irregular notification during symptoms, combined with a normal reading afterward, is a pattern worth reporting even though neither reading alone is conclusive.