Sleep Apnea Feature

DEN230041

Samsung Electronics Co., Ltd · granted 2024-02-06 · product code QZW · Anesthesiology

Premarket evidence — what FDA accepted

Device typesamd
source quote (p.1)
“The Sleep Apnea Feature is an over-the-counter (OTC) software-only, mobile medical application operating on a compatible Samsung Galaxy Watch and Phone.”
Algorithmmachine learned algorithms; performs SpO2 signal interpolation, segmentation, feature extraction, respiratory event classification, and eAHI estimation
source quote (p.3)
“The Sleep Apnea Feature includes machine learned algorithms. The Sleep Apnea Feature evaluates sleep sessions by leveraging the platform's capabilities to acquire PPG signals and derive SpO2 values from those signals. After performing PPG and SpO2 signal quality checks, the algorithm performs this function using 3 steps. 1. Pre-Processing: SpO2 signal interpolation, segmentation, and feature extraction. 2. Respiratory Event Classification: Identify presence of relative SpO2 drop in each 1-minute window. 3. eAHI Estimation and Classification: Enumerate relative SpO2 dips, per-night comparison to the 15 events/hour estimated Apnea/Hypopnea Index (eAHI) threshold.”
Adaptive (vs locked)No
source quote (p.3)
“During their development datasets from representative populations were utilized from over 1000 subjects, split into separate training, tuning, and testing datasets, all maintained independently from the final verification and validation activities.”
PCCPNo
Cybersecurity addressedYes
source quote (p.4)
“Samsung approach to cybersecurity aligns with FDA's 2014 guidance titled, "Content of Premarket Submissions for Management of Cybersecurity in Medical Devices." The device also conforms to the cybersecurity requirements identified in Section 524B to the FD&C Act.”

Validation studies (2)

Bench

sample size not stated

endpoints: SpO2 Data Integrity; Accelerometer Sensor Performance; Sleep Time; On-human Sleep SpO2 Accuracy; On-human Sleep SpO2 Coverage; On-human Stationary SpO2 Accuracy; Low Perfusion

standards: ISO 80601-2-61:2017

Retrospective clinical

n=620 patients

endpoints: differentiating and identifying general population wearable users who show signs of moderate-to-severe obstructive sleep apnea (AHI ≥15) and those users who do not show signs of moderate-to-severe obstructive sleep apnea (AHI <15), as measured by sensitivity and specificity

Reported performance (4 observations)

sensitivity0.827CI [76.7%, 87.6%]
source quote (p.6)
“The Samsung Sleep Apnea Features' sensitivity was 82.7% (167 out of 202 subjects) with a 95% confidence interval of [76.7%, 87.6%], which passes the predetermined acceptance criteria.”
specificity0.877CI [83.1%, 91.4%]
source quote (p.6)
“The DUT's specificity was 87.7% (235 out of 268) with a 95% confidence interval of [83.1%, 91.4%], which did not pass the acceptance criteria.”
agreement_kappaas written: “Single night classification percent agreement”0.842
source quote (p.6)
“The single night classification percent agreement between PSG and DUT is 84.2% (791 out of 930 nights).”
specificityas written: “Modified specificity”0.911CI 95% lower confidence bound of 86.9%
source quote (p.6)
“Considering the benefit received by these 10 subjects, a modified calculation increases specificity to 91.1% (95% lower confidence bound of 86.9%) surpassing the pre-specified specificity acceptance criteria.”

Each value carries its own analysis unit and task — never compare or pool across devices. Source: De Novo decision summary PDF.

Predicate network

Postmarket — what happened after clearance

Not yet tracked — the weekly postmarket refresh hasn't snapshotted this device.

Reimbursement — how devices like this got paid

Not yet tracked — no payment pathway indexed for this clearance (the reimbursement corpus is a growing seed set).

Applicable FDA guidance — what the submission is measured against

FDA guidance documents and guiding principles applicable to De Novo AI/ML devices in the Anesthesiology panel. A curated reference index, not legal or regulatory advice — each item states its own status, and a draft is never binding.

Applicability is derived from the device's FDA advisory panel and pathway — cross-cutting guidances apply to every AI/ML device; panel-specific ones are flagged. Titles, dates, and links verified against fda.gov as of July 2026.

Constat Precedent · public FDA/CMS data · descriptive decision-support, not regulatory or reimbursement advice. Share this page: constat.dev/precedent/device/DEN230041