Study record · validation · 2025

Detection of sleep apnea using only inertial measurement unit signals from Apple Watch: a pilot study with machine learning approach

Hayano J, Adachi M, Murakami Y, and Yuda E

Sleep & Breathing · 2025

Why this study matters to CircaTest

Important because it validates Apple Watch IMU-only sleep apnea detection, which is methodologically distinct from Apple's own sleep apnea notifications feature (which uses combined sensors). Hayano et al. demonstrate that even ACCELEROMETER-ONLY data from the Apple Watch can detect apnea/hypopnea events at AUC 0.831 in a held-out test set. CircaTest cites this when discussing the underlying feasibility of consumer-wearable apnea screening. Caveat: Random Forest models are not the same as Apple's production algorithm; the AUC figure is for the research classifier, not for what an end user sees on a Series 10 or 11.

Abstract

PURPOSE: Despite increased awareness of sleep hygiene, over 80% of sleep apnea cases remain undiagnosed, underscoring the need for accessible screening methods.…

Read the full abstract on the source →

Source: PUBMED · Excerpt for fair-use commentary; full abstract via the source link

Population

Sample size

n = 61

Age

adult

Reference standard

psg

61 adults undergoing polysomnography. 52,337 30-second epochs analyzed; 23.6% identified as apnea/hypopnea episodes. Train/test split 41/20 subjects.

Devices and metrics

Apple Watch (IMU signals only; specific Apple Watch generation not stated in abstract)

All studies for this device →

Metric	Value	95% CI	Note
Accuracy	83.1%	—	Test-set AUC 0.831 for per-epoch respiratory event detection (recorded as percent for display; underlying metric is AUC, not raw accuracy). Random Forest classifier built from extracted seismocardiographic and respiratory signals.
F1 score	0.602	—	Test-set F1 score for per-epoch respiratory event detection.

Apple Watch (IMU signals only; specific Apple Watch generation not stated in abstract)

All studies for this device →

Metric	Value	95% CI	Note
Accuracy	see source	—	Hayano et al. did not specify which Apple Watch generation was used. CircaTest does not extrapolate the AUC 0.831 figure to Series 10 specifically; consult the full paper for the device generation tested.

Cite this study

Hayano J, Adachi M, Murakami Y, and Yuda E (2025). Detection of sleep apnea using only inertial measurement unit signals from Apple Watch: a pilot study with machine learning approach. Sleep & Breathing. https://doi.org/10.1007/s11325-025-03255-w

AI-assembled record · errors are possible

This page was built and is maintained by AI working from the published abstract and PubMed metadata for the cited study. CircaTest does not have access to the full text of every paper in its corpus. Categories of error that have happened on this site and could happen again include:

Per-device metric values inferred when the abstract reports only an aggregate range
Stage-specific accuracy figures pulled from secondary sources rather than the primary paper
Sample size or population characteristics missing from the abstract
Editorial framing in the curatorial note that overstates the study's actual scope
Year typos and author-name transliteration errors

Where a metric value reads see source, the published abstract did not contain that value and CircaTest deliberately left it null rather than guess. For any decision that depends on the underlying numbers, please consult the linked PubMed, PMC, or DOI source directly. Corrections are welcome via the about page.

Source links

Added to the CircaTest meta-analysis on 2026-04-06. How CircaTest evaluates studies →