Apple Watch breathing alerts send a 79-year-old to CPAP therapy

Fukuoka case (Cureus, 26 May 2026): sleep-apnoea notifications prompted consultation; watch ECGs showed AF only after “elevated” breathing-disturbance nights; CPAP under way, no polysomnography.

1ⓘ

AI system

System

Apple Watch Sleep Apnea Notification Feature (breathing-disturbance notifications) and the ECG app

Model

Not assessed

Release

Exact release not established.

Application developer

Apple

Role

The model detected elevated breathing disturbance and generated a notification.

Configuration

The participant used an Apple Watch Series 11 with an iPhone 15 Pro.

Inputs and outputs

Not assessed

Autonomy

Not assessed

Workflow

Nightly watch wear with the sleep-apnea notification feature active, plus the owner's habit of recording a watch ECG every morning and evening. A known history of paroxysmal AF that made the morning traces interpretable to his cardiologists. Access to a clinic willing to act on watch data (CPAP started without polysomnography).

Human oversight

Not assessed

Relation to AI

The notification feature's deep-learning model, described in Apple's submission to FDA, flagged elevated breathing disturbance and brought the patient to the clinic. His own morning ECG recordings, read by the authors, showed the AF paroxysms. The paper does not state separately what the ECG app's classifier reported. Clinicians chose CPAP and deferred ablation. The treating authors report the clinical episode. Apple supplies the technical description and validation in its 510(k) summary. FDA's clearance is a separate regulatory decision and does not verify this patient's benefit. [1] [2]

AI contribution

Sleep-apnea notifications prompted the patient to seek care. Clinicians interpreted his morning ECG recordings and chose CPAP treatment. The report does not separately identify an ECG classifier’s contribution or provide a treatment outcome. [1] [2]

Benefits

Outcome identifiers

Not assessed

Benefit

Fukuoka case (Cureus, 26 May 2026): sleep-apnoea notifications prompted consultation; watch ECGs showed AF only after “elevated” breathing-disturbance nights; CPAP under way, no polysomnography.

Category

Improved Physical Health

Status

Not assessed

Beneficiary groups

People.

Beneficiaries

Not assessed

Reach

Not assessed

Distribution and access

Not assessed

Duration

Not assessed

Expected?

Yes

Costs and harms

Not assessed

Measurement

Studies

Not assessed

Study design

Not assessed

Measurement

Not assessed

Setting and sample

Not assessed

Measures

Not assessed

Comparison

ambulatory monitoring and standard 12-lead ECG, which had documented no paroxysms since his retirement

Results

Reference: AF paroxysms were not documented by ambulatory monitoring or standard ECG. Result: AF paroxysms were detected repeatedly in the morning after notification of nocturnal “elevated” breathing disturbance, but not in the morning after nights when breathing disturbance was “not elevated”.

Additional value from AI

Not assessed

Limitations

Not assessed

Denominators

Not assessed

Materials for replication

Not assessed

Money

Kind of value

Not assessed

Amount

Not assessed

Currency and prices

Not assessed

Period and unit

Not assessed

Realization

Not assessed

Recipients and payers

Not assessed

Gross or net

Not assessed

Included costs

Not assessed

Comparison

Not assessed

Method

Not assessed

Return and payback

Not assessed

People and places

Participants

Developer: Apple.

Jurisdictions

Benefit: Fukuoka, Japan. Developer: United States.

Ownership and control

Not assessed

Dependencies

Not assessed

Cross-border connections

Not assessed

Geopolitical implications

Not assessed

Follow-up

Conditions for replication

The user completes onboarding and wears the watch during sleep, since the feature evaluates 30-day windows opportunistically. Clinicians confirm with conventional ECG and polysomnography, which the authors say the watch cannot replace and which was not done here.

Replication attempts

Not assessed

Later findings

The May 2026 report says CPAP therapy is under way and ablation will depend on its effect; no later outcome was found.

Availability

Not assessed

Institutional decision

Not assessed

Grounds

Not assessed

Action and result

Not assessed

Success criterion

Not assessed

Evidence

Source numbers

AIBD-0300-source-0001–0002

Citations

Numbered references below. [1] [2]

Language

Not assessed

Source relationships

Independent, Developer [1] [2]

Original evidence

Treating clinicians report the consultation and ECG findings. Apple’s FDA submission describes the notification algorithm and its validation. FDA’s clearance is a regulatory decision, not confirmation of this patient’s benefit. [1] [2]

Access and versions

The clinical report gives no sleep-study confirmation or treatment response and does not identify the ECG classifier’s contribution. Nightly breathing-disturbance ratings differ from the feature’s 30-day notification. Apple’s underlying training and validation records are not public. [1] [2]

Supported findings

Not assessed

Evidence strength

Not assessed

Missing evidence

Not assessed

Grades

Assessed benefits

People.

Depth

People: 1.

Reasons

People: A man whose AF was thought quiescent learned that paroxysms recur after disturbed-breathing nights, and he started CPAP. No confirmed sleep-apnea diagnosis and no treatment effect are reported.

Observed or intended

Not assessed

Rubric

Not assessed

Assessor and date

Not assessed

Alternative assessments

Not assessed

Fill

Independently studied [1] [2]

Entry details

Kind

Not assessed

Field

Healthcare (cardiology; sleep medicine). Consumer wearables.

Purpose

The notification feature is built to identify patterns of breathing disturbance suggestive of moderate-to-severe sleep apnea and tell the user. The ECG app is built to record and classify single-lead rhythm. Both were used as designed.

Dates

Source published: 26 May 2026. Updated: 24 October 2026. [1] [2]

Stage

Not assessed

Version

SHA-256: 3c00ba05fd9e…

References

Other identifiers

Not assessed

Category mappings

Not assessed

Brief

Not assessed

Revisions

6 revisions. Latest: 24 October 2026. [1] [2]

Sources

  1. Case report Independent author

    Maruyama T, Hieda M, Fukata M (2026). Smartwatch-Detected Atrial Fibrillation on the Morning After a Nocturnal Sleep-Disordered Breathing Notification: A Case Report. Cureus 18(5):e109701 (PMC13301138). Case Presentation, Figure 2, Discussion, and references 12–14.

    Original clinical narrative, ECGs and breathing-disturbance chart document consultation and management. Case Presentation; Figure 2; Discussion; references 12–14.

    Treating clinical authors report the episode and declare no relevant support or financial relationships.

  2. Regulatory document Developer

    Apple Inc. (2024). Sleep Apnea Notification Feature: 510(k) Summary, K240929, device description and performance testing, pp. 2 and 5–7, filed with the FDA. File includes FDA's September 13, 2024 substantial-equivalence letter.

    Manufacturer specifies the deep-learning model, training and 30-day notification rule; FDA letter establishes clearance. FDA letter pp. 1–3; Apple summary pp. 2 and 5–7.

    Mixed file: FDA's decision plus Apple's technical submission. The technical claims are maker-authored.

Collections: Scrutiny and findings, Cross-border cases.