Samsung's Sleep Apnea Check Now Grades Severity, Not Just Signs

Last updated: October 2026 · Based on Samsung's October 28, 2026 announcement, its July 2026 U.S. FDA clearance for the upgraded feature, and the original feature's 2024 FDA De Novo authorization.

For two years, a Galaxy Watch could answer one sleep question with a yes or a no: did it see signs consistent with moderate to severe obstructive sleep apnea, or did it not? On October 28, 2026, Samsung announced an upgraded version that tries to answer a second question — how often the disruptions are happening. Instead of a single flag, the new feature counts breathing disruptions per hour and sorts the result into three levels. That is a meaningful step for a wrist device, and also an easy one to overread. A graded screening result is still a screening result, not a diagnosis.

What Samsung announced on October 28, 2026

Samsung said the upgraded feature had been approved by South Korea's Ministry of Food and Drug Safety, following clearance in the United States by the Food and Drug Administration in July 2026. Reporting on the FDA record identifies the upgraded version as Sleep Apnea Feature v2.0, cleared through the 510(k) pathway on July 20, 2026. The rollout starts in South Korea in November 2026, first on the Galaxy Watch Ultra 2, Galaxy Watch 9 and Galaxy Ring, and then on the Galaxy Watch 7 and Galaxy Watch 8 through software updates for devices running Wear OS 7.0 or later.

Two limits on that news matter for readers outside Korea. First, Samsung has confirmed the November start for Korea only; it has not announced a United States or European date for the upgraded version, even though the U.S. clearance is already in place. Second, this is not a replacement for every Galaxy owner. The original feature — the one that only reports whether signs were detected — stays available on Galaxy Watch 4 and newer models running Wear OS 5.0 or later. If your watch is not on the upgrade list, you keep the older yes-or-no version, not nothing.

How the grading works

The feature runs through the Samsung Health Monitor app on a Galaxy device. You record at least five hours of sleep on two occasions within ten days, and the system analyses those nights together. An AI algorithm estimates how many breathing disruptions occur per hour, and the result is grouped into three levels based on frequency, described in Samsung's materials as mild, occasional and frequent.

Underneath that simple output is a sensor change worth understanding. Samsung's original feature, first approved in Korea in October 2023 and given FDA De Novo authorization in February 2024 as a new over-the-counter device category for assessing sleep apnea risk, relied mainly on blood-oxygen patterns measured optically during sleep. Reporting on v2.0 describes a broader input set, combining optical photoplethysmography signals with movement data. In plain terms, the watch is no longer looking at oxygen dips alone; it is cross-checking them against how the body moves. That broader input is also part of why the feature can now reach the Galaxy Ring, a smaller device that is often more comfortable to wear overnight — and a device you do not wear overnight gathers no sleep data at all.

How this differs from Apple's approach

Apple Watch takes a different route to a related question. Apple's sleep apnea notifications, cleared in September 2024, use the accelerometer to detect wrist-movement patterns linked to interrupted breathing, and they judge a much longer window — 30 days of nighttime data — before producing a result, grouped as elevated or not elevated. Samsung asks for two qualifying nights inside ten days and now returns three frequency levels.

Neither approach is a sleep study, and the difference in timing cuts both ways. A two-night window is faster and easier to complete, but sleep varies from night to night, so a short window can catch an unusually bad — or unusually good — stretch. A 30-day window smooths some of that variation and takes far longer to say anything. Our AI smartwatches guide places these cleared screening features alongside ordinary wellness sleep scores, which carry no clearance at all, and our explainer on Apple Watch hypertension alerts makes the same core point for a different condition: a notification is a prompt to measure properly, not a measurement itself.

What the evidence does — and does not — show

Samsung states the feature is intended for adults aged 18 and over who have not already been diagnosed with sleep apnea, and that it does not replace professional diagnosis or consultation. Treat that wording as part of the product, not as fine print. The clinical reference standard remains an overnight sleep study — in a lab, polysomnography, or a prescribed home sleep apnea test — which can measure airflow, oxygen, heart activity and sleep stages directly in ways a wrist or finger device estimates indirectly.

Independent testing of the earlier Galaxy Watch feature shows why confirmation matters. A validation study discussed in the Journal of Clinical Sleep Medicine in 2026 compared the watch's estimated apnea-hypopnea index against in-lab polysomnography in 147 participants over two nights. At the manufacturer's default threshold, the watch was good at catching cases but produced a notable share of false positives; moving the threshold higher reversed that trade-off. That study tested the earlier feature, not v2.0, so its exact numbers should not be pasted onto the new grading. Its lesson does transfer: any single cutoff on a wearable will miss some people and flag some people who would not be confirmed on a full study. A three-level result does not remove that trade-off — it gives you more context inside it.

Samsung also cites U.S. National Sleep Foundation figures that about 25% of men and 10% of women in the United States have sleep apnea, many of them unaware of it. That prevalence is the reason screening exists at all. A tool that reaches people who would never book a sleep study can do real good, provided its results start a clinical conversation instead of ending one.

What to do with each kind of result

If your result is in the highest level: book the clinical follow-up. Take the app result, the dates, and any symptoms you or a partner have noticed — loud snoring, witnessed pauses, morning headaches, daytime sleepiness — to a clinician, who can decide whether a sleep study is warranted. Do not buy a treatment device on the strength of a watch grade.

If your result is in a lower level: read it as information, not reassurance. A mild or occasional grade on two nights does not rule out apnea on other nights, in other sleep positions, or after alcohol or illness. If symptoms persist, the symptom history matters more than a single low grade.

If you get no flagged signs: the same caution applies as with every wearable screen. The feature was built around signs of moderate to severe apnea in undiagnosed adults. It was not built to clear anyone, and it does not assess every sleep problem.

If you already have a diagnosis: this feature was not built or cleared for managing your condition. Follow your existing care plan, and treat any watch output as, at most, something to mention at your next review.

Should this feature decide your next watch?

Only if you will actually wear the device overnight, with charge to spare, on a Galaxy phone setup that supports Samsung Health Monitor in your country. Those practical conditions eliminate more buyers than the sensor list does. If night comfort is your sticking point, the Ring's inclusion in the first wave is the most interesting part of this announcement — a lighter device you keep on is worth more than a more capable watch sitting on a charger. Our first watch guide puts fit and daily wear ahead of feature lists for exactly that reason, and our guide to smartwatch battery health covers the overnight-wear habit that makes any sleep feature usable at all: a watch that needs charging at bedtime will never collect two qualifying nights.

Wait for your region's date before paying a premium for v2.0: as of October 31, 2026, the confirmed start is Korea in November. A U.S. clearance is a necessary step, not a launch date.

Sources

Samsung announcement, October 28, 2026, reported by GSMArena, GHacks and the Seoul Economic Daily: upgraded feature, AI count of breathing disruptions per hour, three levels, two nights of at least five hours within ten days, Korea-first November rollout (Ultra 2, Watch 9, Ring; Watch 7 and Watch 8 later on Wear OS 7.0 or later), original feature retained on Watch 4 and later on Wear OS 5.0 or later, availability of the original feature in 84 countries.
FDA record reporting by MobiHealthNews and TechTimes, October 2026: Sleep Apnea Feature v2.0 510(k) clearance dated July 20, 2026; intended for undiagnosed adults; not a diagnosis or replacement for professional evaluation; original feature's FDA De Novo authorization in February 2024 following Korea approval in October 2023.
Journal of Clinical Sleep Medicine validation discussion, 2026: 147-participant comparison of the earlier Galaxy Watch feature against in-lab polysomnography, with sensitivity and specificity trading off by threshold — cited here for the screening trade-off, not as v2.0 performance figures.

Go deeper with AI

This article is the starting point. Copy any of these prompts into your favorite AI assistant to learn more about Samsung's graded sleep apnea screening:

🔎 Turn a graded result into a clinician-ready note

My Samsung Health Monitor sleep apnea result was [mild / occasional / frequent] from two nights on [dates]. I am [age], have not been diagnosed with sleep apnea, and I notice [snoring / witnessed pauses / morning headaches / daytime sleepiness / none of these]. Write a short note I can take to a clinician: what the wearable result does and does not show, what a sleep study would add, and which questions I should ask about whether I need one. Flag that this is general information, not medical advice.

🧭 Compare Samsung and Apple sleep apnea screening fairly

Compare Samsung's upgraded sleep apnea feature (two nights of at least five hours within ten days, three frequency levels, FDA-cleared July 2026, Korea-first rollout) with Apple Watch sleep apnea notifications (accelerometer-based, 30-day window, elevated or not elevated). For each, state what signal it uses, how long it takes, who it is intended for, and what confirmation step should follow a flagged result. Do not treat either as a diagnosis.

💡 Check whether you will actually complete the test

I own [Galaxy Watch / Ring model] and a [phone model] in [country]. Based on Samsung's October 2026 rollout (v2.0 first on Ultra 2, Watch 9 and Ring in Korea in November; Watch 7 and 8 later on Wear OS 7.0 or later; original feature on Watch 4 and later), tell me which version I can use now, what I still need to confirm for my country, and a realistic two-night plan — charging, fit and app steps — to record two nights of at least five hours within ten days.

Tip: replace the bracketed parts with your own situation — the more specific your prompt, the more useful the answer.

Summary

Samsung's upgraded sleep apnea feature, FDA-cleared in July 2026 and approved in Korea on October 28, 2026, moves from a yes-or-no flag to an AI estimate of breathing disruptions per hour, graded as mild, occasional or frequent from two nights of at least five hours within ten days. It launches first in Korea in November on the Galaxy Watch Ultra 2, Galaxy Watch 9 and Galaxy Ring, with Watch 7 and Watch 8 to follow by update; no U.S. launch date had been announced as of October 31, 2026, and the original feature remains on Watch 4 and later. It is a screening tool for undiagnosed adults, not a diagnosis.

The single most important takeaway: a three-level grade gives a flagged result more context, but every level still ends in the same place — a proper sleep study is what confirms or rules out sleep apnea.

Next steps: read our AI smartwatches guide for how health features compare, or start with our first watch guide for sizing and buying safely.