Overnight SpO₂ Monitoring for Sleep Apnea: How Wearables Detect Hidden Breathing Disorders

Overnight SpO₂ monitoring helps detect hidden breathing disorders such as obstructive sleep apnea by continuously tracking blood oxygen levels during sleep. When breathing repeatedly slows or stops, oxygen levels temporarily drop, creating characteristic desaturation patterns that wearable devices can detect. Modern smartwatches, rings, and wearable oximeters use optical sensors and AI-powered analysis to identify these overnight oxygen fluctuations, helping flag individuals who may be at risk of sleep apnea. Clinical studies have shown that wearable SpO₂ monitoring can achieve high sensitivity for detecting sleep-disordered breathing, making it a promising tool for screening, long-term monitoring, and identifying sleep-related respiratory problems that often go unnoticed during the day. 

---

Sleep apnea is one of the most underdiagnosed sleep disorders in the world. Millions of people experience repeated breathing interruptions during sleep without realizing it. Symptoms such as fatigue, poor concentration, morning headaches, and snoring are often dismissed as normal consequences of stress, aging, or lifestyle habits. 

The challenge is that these breathing disturbances occur when a person is asleep, making them difficult to observe without monitoring. This is where overnight SpO₂ tracking has become increasingly valuable. By continuously measuring blood oxygen levels throughout the night, modern wearables can help identify patterns that may indicate hidden breathing disorders long before they are formally diagnosed.

AI generated image

Sleep Apnea and Oxygen Desaturation

Obstructive sleep apnea is extremely common. An estimated 936 million adults worldwide (ages 30–69) have OSA, yet the vast majority are unaware.¹ In fact, over 80% of people with moderate-to-severe OSA go undiagnosed.² Untreated OSA raises risks of hypertension, heart disease, stroke and daytime accidents. One reason it is so often “hidden” is that apneas occur only during sleep.

AI generated image

However, a hallmark of sleep apnea is repeated oxygen dips. Each time the airway partially or fully collapses, blood oxygen (SpO₂) falls sharply. While these desaturations are invisible to the patient, they do show up on pulse-oximeter recordings. For decades, doctors have used overnight pulse oximetry as an economical screening tool for sleep apnea. As one review noted, “overnight pulse oximetry helps determine the severity of disease and is used as an economical means to detect sleep apnea”.³ In other words, frequent SpO₂ drops during sleep strongly suggest undiagnosed OSA.

Which Types of Sleep Apnea Can Wearables Help Detect?

Sleep apnea is not a single condition. It is generally classified into three main types:

• Obstructive Sleep Apnea (OSA): The most common form, caused by repeated collapse of the upper airway during sleep.

• Central Sleep Apnea (CSA): Occurs when the brain temporarily fails to send the proper signals to the muscles that control breathing.

• Complex (Treatment-Emergent) Sleep Apnea: A combination of obstructive and central events, usually seen after treatment for OSA has begun.

Most consumer wearables are designed primarily to screen for obstructive sleep apnea, since repeated airway blockages produce the characteristic drops in blood oxygen (SpO₂), heart rate fluctuations, and breathing disruptions that wearable sensors can detect.

Detecting central sleep apnea is much more challenging. Although oxygen levels may also fall, SpO₂ alone cannot determine whether breathing stopped because the airway collapsed or because the brain temporarily paused the breathing signal. Making that distinction requires additional physiological measurements, such as respiratory effort and brain activity, which consumer wearables generally cannot capture.

How Wearable Devices Detect Sleep Apnea Using Overnight SpO₂ Monitoring

Traditional pulse oximetry was once limited to fingertip or earlobe sensors used in hospitals and clinics. Today, many wearable devices including smartwatches, smart rings, fitness bands, and even chest patches, can continuously measure blood oxygen saturation (SpO₂) throughout the night using photoplethysmography (PPG).

PPG works by shining light into the skin and measuring how much light is absorbed or reflected by blood vessels with each heartbeat. Because oxygen-rich and oxygen-poor blood absorb light differently, the sensor can estimate your blood oxygen level continuously while you sleep.

Over the course of the night, these devices build a detailed record of your SpO₂ levels. The real value, however, lies not in individual readings but in the overall pattern. During an obstructive sleep apnea event, the airway temporarily collapses, reducing airflow into the lungs. As a result, blood oxygen levels begin to fall before recovering once normal breathing resumes. When this cycle repeats dozens or even hundreds of times during the night, it creates a characteristic pattern of repeated oxygen desaturations.

To identify these events, wearable devices use algorithms that first clean the raw data by removing noise caused by movement or temporary sensor dropouts. They then analyze the signal for recurring oxygen drops and calculate metrics such as the Oxygen Desaturation Index (ODI), i.e., the number of times per hour that blood oxygen falls by at least 3–4%.

ODI is one of the most important indicators used in sleep apnea screening because it closely correlates with the Apnea-Hypopnea Index (AHI), the clinical measure used to determine sleep apnea severity during a sleep study. Numerous studies have shown that ODI can be used to estimate AHI with good agreement⁴, making it a valuable screening tool when a full sleep study is not immediately available.

However, modern wearables don't rely on ODI alone. They also analyze other overnight oxygen metrics, including the average SpO₂, the lowest oxygen level reached during sleep (oxygen nadir), the amount of time spent below clinically important thresholds such as 90% saturation, and how quickly oxygen levels recover after each desaturation. Some devices also combine these oxygen patterns with heart rate, movement, and respiratory signals collected from the same PPG sensor to improve detection accuracy.

Many newer systems further enhance this analysis using artificial intelligence and machine learning. Rather than looking at individual oxygen drops in isolation, these algorithms evaluate the entire overnight SpO₂ waveform to recognize breathing patterns that are strongly associated with obstructive sleep apnea. This combination of continuous sensing and intelligent analysis allows wearable devices to identify people who may have previously undiagnosed breathing disorders while they sleep.

Evidence: How Accurate Are Wearable Oximeters?

Several recent studies have validated wearable SpO₂ monitors against lab sleep studies (polysomnography, PSG) or standard home sleep tests. The results are encouraging:

In one prospective validation study, researchers tested an AI algorithm that analyzed only overnight SpO₂ data collected from a wearable device in 53 people suspected of having obstructive sleep apnea (OSA). Even without measuring breathing effort or brain activity, the system correctly classified the severity of OSA with 89% accuracy. By comparison, traditional home sleep apnea tests (HSATs) have reported accuracies of around 61% in similar settings.⁵ This suggests that AI-powered analysis of overnight blood oxygen patterns has the potential to match, and in some cases outperform, conventional home-based screening methods.

In another clinical study involving 102 people who underwent both a sleep lab test (polysomnography) and a wrist-worn pulse oximeter, the wearable performed remarkably well. The results from the device closely matched the sleep lab findings (correlation r = 0.87). When screening for sleep apnea, it correctly identified over 90% of people who had the condition (90.7% sensitivity) and produced no false positives in that study (100% specificity). Overall, the device demonstrated excellent screening accuracy (AUC = 0.994), suggesting that wearable pulse oximeters can be highly effective at identifying people who should undergo further clinical evaluation.⁶

A 2024 systematic review and meta-analysis evaluated 38 studies on AI-enabled wearable devices, including pulse oximeters, ECG wearables, smartwatches, and other sensor-based systems. Across these studies, wearable technologies achieved a pooled accuracy of 86.9%, sensitivity of 93.8%, and specificity of 75.2% for detecting sleep apnea.⁷ In practical terms, this means wearables were highly effective at identifying people who truly had sleep apnea, although some individuals without the condition were incorrectly flagged. The authors concluded that wearable AI shows strong potential for identifying and classifying sleep apnea, but should currently be used alongside conventional diagnostic methods rather than replacing them.

These results indicate that continuous SpO₂ monitoring, especially when combined with smart algorithms, can indeed flag hidden sleep apnea in high-risk people. 

Practical Benefits of Wearable SpO₂ Screening

For patients and clinicians, continuous oximetry unlocks several advantages:

• Accessibility: Instead of an expensive sleep lab, patients can wear a lightweight device at home. This can greatly expand access to screening.

• Continuous data: Wearables can record many nights of data unobtrusively. This captures night-to-night variability and can identify cases that a single-night study might miss.

• Early warning: For people with cardiovascular risk factors or symptoms (snoring, daytime sleepiness), a wearable SpO₂ recorder could raise an alert to see a sleep doctor. It essentially turns every normal person’s smartwatch into a rough sleep screen.

• Monitoring treatment: Wearable oximetry can also track the response to therapy (CPAP or oral appliance) at home by showing improvement in overnight oxygen stability.

However, it’s important to note that not all wearables are medical-grade. Consumer devices vary in accuracy, and motion artifacts can distort readings. High-quality, FDA-cleared wearable oximeters are best for clinical decisions. But even off-the-shelf devices can provide useful clues when interpreted cautiously.

Limitations of Wearable Apnea Monitoring

It’s important to recognize limitations. 

A wearable SpO₂ monitor is not a definitive diagnosis of sleep apnea on its own. False positives can occur (e.g. poor sensor contact or other lung issues can mimic desaturations). Mild OSA may also be missed if desaturations are subtle. Moreover, some sleep apneas, especially purely upper airway resistance without much oxygen drop, might not trigger clear SpO₂ events.

Current medical guidelines still consider polysomnography (PSG) or multi-channel home sleep testing the standard methods for diagnosing sleep apnea. While wearable devices can be excellent screening tools, they are not yet reliable enough to replace these clinical tests. If a wearable detects signs of possible sleep apnea, the next step should always be a comprehensive sleep evaluation.

Future Directions

The field is advancing rapidly. As AI models improve and sensors get better, wearable SpO₂ screening will likely become more accurate. For instance, Apple’s latest sleep-app update claims to notify users of potential sleep apnea by analyzing nocturnal breathing disturbances (using accelerometer and heart-rate data).⁸ Similarly, research efforts like SleepAI⁹ and other apps are poised to leverage nights of SpO₂ data to predict apnea events in real-time.

Eventually, we may see turnkey solutions: imagine a watch that automatically computes your ODI and flags “likely moderate sleep apnea” if it stays elevated for multiple nights. Such tools could prompt earlier diagnosis and treatment, helping to “minimize the negative health impacts” of OSA around the globe.

Conclusion

Sleep apnea is a hidden epidemic, affecting nearly a billion people worldwide, and overtreatment requires finding the sleepers who wake gasping for air. Overnight SpO₂ monitoring offers a powerful window into breathing during sleep, and wearable oximeters and smartwatches are turning this concept into reality. By continuously logging your blood oxygen, they can reveal the telltale dips of undiagnosed apnea. As studies show, the combination of wearable SpO₂ data and AI algorithms can detect sleep apnea with high accuracy. While not a substitute for a formal sleep study today, wearable SpO₂ monitoring is a valuable screening tool that can bring sleep-disordered breathing out of the shadows.

---

References:

1. Iannella G, Pace A, Bellizzi MG, Magliulo G, Greco A, De Virgilio A, Croce E, Gioacchini FM, Re M, Costantino A, Casale M, Moffa A, Lechien JR, Cocuzza S, Vicini C, Caranti A, Marchese Aragona R, Lentini M, Maniaci A. The Global Burden of Obstructive Sleep Apnea. Diagnostics (Basel). 2025 Apr 25;15(9):1088. doi: 10.3390/diagnostics15091088. PMID: 40361906; PMCID: PMC12071658.

2. Faria A, Allen AH, Fox N, Ayas N, Laher I. The public health burden of obstructive sleep apnea. Sleep Sci. 2021 Jul-Sep;14(3):257-265. doi: 10.5935/1984-0063.20200111. PMID: 35186204; PMCID: PMC8848533. 

3. Netzer N, Eliasson AH, Netzer C, Kristo DA. Overnight pulse oximetry for sleep-disordered breathing in adults: a review. Chest. 2001 Aug;120(2):625-33. doi: 10.1378/chest.120.2.625. PMID: 11502669. 

4. Sharma, P., Thakur, S., Rai, D. K., Karmakar, S., & H, A. (2024). Connecting the dots: Analysing the relationship between ahi and ODI in obstructive sleep apnea. Sleep Science and Practice, 8(1). https://doi.org/10.1186/s41606-024-00102-x 

5. Attia, S., Oksenberg, A., Levy, J., Ang, A., Shani‐Hershkovich, R., Adler, A., Katsav, S., Haimov, S., Alexandrovich, A., Tauman, R., & Behar, J. A. (2025). Clinical validation of artificial intelligence algorithms for the diagnosis of adult obstructive sleep apnea and sleep staging from oximetry and photoplethysmography—sleepai. Journal of Sleep Research, 35(1). https://doi.org/10.1111/jsr.70093 

6. Jabbaripour, S., Saien, S., Heidari, R., Erfanian, R., & Amali, A. (n.d.). Performance of Wrist-Worn Pulse Oximeter for the Screening of Obstructive Sleep Apnea. Iranian Journal of Otorhinolaryngology . https://doi.org/https://doi.org/10.22038/ijorl.2024.77290.3586 

7. Abd-Alrazaq A, Aslam H, AlSaad R, Alsahli M, Ahmed A, Damseh R, Aziz S, Sheikh J. Detection of Sleep Apnea Using Wearable AI: Systematic Review and Meta-Analysis. J Med Internet Res. 2024 Sep 10;26:e58187. doi: 10.2196/58187. PMID: 39255014; PMCID: PMC11422752. 

8. Groundbreaking health features available today on Apple Watch and airpods pro - apple (in). Apple Newsroom. (2026, May 21). https://www.apple.com/in/newsroom/2026/05/groundbreaking-health-features-available-today-on-apple-watch-and-airpods-pro/ 

9. Sleepai. (n.d.-d). https://sleep-ai.org/