The marquee health feature in this year's Apple Watch Series 10 is sleep apnea detection, although it's not exclusive to the new model -- it's also coming to last year's Apple Watch Series 9 and Apple Watch Ultra 2 as part of watchOS 11.

However, when Apple announced the new sleep tracking feature during last Monday's Glowtime event, it wasn't entirely clear when that would happen. During the presentation, Dr. Sumbal Desai, Apple's VP of Health, noted that the company was still waiting for clearance from the US Food and Drug Administration (FDA) before launching it in the US, although she added that Apple expected that to arrive "soon."

As with the new Hearing Aid feature on the AirPods Pro 2, Dr. Desai wasn't merely being optimistic. The FDA has approved both these features in record time, and sleep apnea detection has been given the green light to roll out on the two 2023 Apple Watch models as part of today's watchOS 11 update. It will also be ready to go when the Apple Watch Series 10 lands in stores (and customers' eager hands) later this week. Technically, the FDA only lists an "Over-the-counter device to assess risk of sleep apnea," but there's no other product Apple makes that fits that description.

While these approvals may have seemed quick, Apple has had the applications in the works for some time. Apple applied for a "Sleep Apnea Notification Feature (SANF)" on April 4, 2024. If anything, it probably asked the FDA to delay publishing its approvals to avoid tipping its hand before its Glowtime event; the approval date of September 13 can't be merely a coincidence.

Either way, it's good news for owners of recent Apple Watch models -- and anyone planning to pick up a new Series 10 later this week. What's more, the ongoing patent dispute that's blocked the blood oxygen sensor for US models won't affect the new feature as Apple isn't relying on that particular sensor to work its magic.

For those interested in a deeper dive, Apple has published a new white paper outlining the feature, titled Estimating Breathing Disturbances and Sleep Apnea Risk from Apple Watch. The paper explains how Apple is using an accelerometer-based breathing analysis to detect patterns typical of sleep apnea:

Apple Watch tracks movement with triaxial accelerometer signals, which capture coarse motion of the body as well as fine movements including motion associated with breathing. Apple developed an algorithm that uses the accelerometer time series data to classify Breathing Disturbances that occur during sleep tracking, which are temporary interruptions in the breathing pattern.

As Dr. Desai explained during Apple's Glowtime event, this is backed by a machine learning algorithm developed and validated against extensive clinical test data. The paper goes into more detail:

Apple conducted research studies to develop the Breathing Disturbances metric and the associated sleep apnea notification algorithm. Adult participants from multiple research sites provided informed consent via protocols approved by an institutional review board (IRB). To enhance performance generalizability, Apple recruited a diverse population of research participants across various demographic factors (age, biological sex, race, ethnicity, and BMI) and evaluated sleep in both at-home and in-laboratory sleeping environments. In addition, the studies included a broad range of sleep apnea severity, from normal (fewer than five apnea and hypopnea events per hour) to severe (more than 30 events per hour).

To put it all together, Apple had a group of research participants wear an Apple Watch while also undergoing polysomnography (PSG) tests in a lab and home sleep apnea test (HSAT) kits at home to calibrate the results against the American Academy of Sleep Medicine (ASAM) standards. A total of 3,936 nights of recordings were collected from 2,160 participants, with some contributing multiple nights of recordings. Once the algorithm was designed, Apple then tested its results against an additional 7,220 nights of recordings from 2,542 participants, using an entirely new set of data to validate the algorithm's output.

The finished algorithm was then formally validated in a separate clinical study for submission to the FDA and other global health regulators. The clinical study had 1,499 participants wear an Apple Watch for up to 30 nights plus take at least two nights of HSAT recordings for comparison. The results determined that the Apple Watch showed accurate results with no false positives. All participants with a positive algorithm result had at least mild sleep apnea, and 89.1% of severe sleep apnea cases were correctly identified. Sensitivity was lowest in the "moderate" category of sleep apnea, but this only suggests that the level of sleep apnea is more likely to be misidentified as "normal" or "mild" rather than missed entirely.

The FDA determined that Apple's solution meets the standards for sleep apnea tracking devices as it's "substantially equivalent" to other methods already approved by the agency.