Search Results

The Sleep Apnea Notification Feature (SANF) is a software-only mobile medical application that analyzes Apple Watch sensor data to identify patterns of breathing disturbances suggestive of moderate-to-severe sleep apnea and provides a notification to the user. This feature is intended for over-the-counter (OTC) use by adults age 18 and over who have not previously received a sleep apnea diagnosis and is not intended to diagnose, treat, or aid in the management of sleep apnea. The absence of a notification is not intended to indicate the absence of sleep apnea.

The Sleep Apnea Notification Feature (SANF) is an over-the-counter mobile medical application (MMA) intended to identify patterns of breathing disturbances suggestive of moderate-to-severe sleep apnea and provide a notification to the user. SANF is intended to run on compatible iOS (e.g. iPhone, iPad) and Apple Watch platforms. Users set up SANF and view their health data on the iOS platform. Prior to use, users must undergo educational onboarding. SANF uses accelerometer sensor data collected by the Apple Watch to calculate breathing disturbance values while a user is asleep. Breathing disturbances describe transient changes in breathing patterns, such as temporary breathing interruptions.

Breathing disturbance data is analyzed in discrete, consecutive 30-day evaluation windows, If patterns consistent with moderate-to-severe sleep apnea are identified within the 30-day evaluation window, the user is notified. SANF provides visualizations depicting the user's breathing disturbance data over various time scales. SANF is not intended to provide instantaneous measurements. Instead, once activated, SANF runs opportunistically in the background receiving signals from Apple Watch sensors for processing.

Here's a summary of the acceptance criteria and study details for the Sleep Apnea Notification Feature (SANF), based on the provided FDA 510(k) summary:

1. Table of Acceptance Criteria and Reported Device Performance

Note: The document emphasizes that performance was "optimized for high specificity" given the opportunistic detection nature of the device. This implies that while a specific numerical sensitivity might not have been a hard "acceptance criterion" per se, the reported sensitivity alongside high specificity demonstrated sufficient effectiveness for clearance.

2. Sample Size Used for the Test Set and Data Provenance

• Sample Size for Test Set (Clinical Study):

  • Notification Performance Analysis: 1,278 subjects
  • Breathing Disturbance Performance Analysis: 1,305 subjects
  • Total Subjects Enrolled: 1,499 subjects (some had insufficient data for analysis)

• Data Provenance:

  • Country of Origin: United States (from "several sites across the United States").
  • Retrospective or Prospective: Prospective. The study "enrolling 1,499 subjects" suggests a prospective collection of data specifically for this validation study.

3. Number of Experts Used to Establish the Ground Truth and Qualifications

The document refers to the "Nox T3s home sleep apnea testing (HSAT) device (K192469) as a reference device" for ground truth. The HSAT device itself provides the AHI (Apnea-Hypopnea Index) which is the clinical standard for sleep apnea diagnosis.

Thus, the ground truth was established by the HSAT device, not by human experts directly adjudicating each case. The output of the HSAT device is the ground truth measure (AHI).

4. Adjudication Method for the Test Set

The ground truth was established by the Nox T3s HSAT device, which is an objective measurement device. Therefore, a human expert adjudication method (like 2+1 or 3+1) was not explicitly mentioned or performed for the primary clinical endpoint, as the HSAT device is considered the reference standard. The AHI values derived from the HSAT device served as the diagnostic ground truth.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

There is no mention of an MRMC comparative effectiveness study involving human readers with or without AI assistance. The study focuses on the standalone performance of the SANF device against a reference standard (HSAT).

6. Standalone (Algorithm Only Without Human-in-the-Loop Performance)

Yes, a standalone performance study was done. The reported sensitivity and specificity values are for the algorithm's performance in identifying patterns suggestive of moderate-to-severe sleep apnea and providing a notification, without human intervention in the interpretation or decision-making process based on the device's output. The device itself "provides a notification to the user," implying direct algorithm output.

7. Type of Ground Truth Used

The ground truth used was objective diagnostic data derived from a medical device: The Nox T3s home sleep apnea testing (HSAT) device, which provides the Apnea-Hypopnea Index (AHI). This is considered a gold standard for diagnosing and classifying the severity of sleep apnea in a home setting.

8. Sample Size for the Training Set

The algorithm development dataset included over 11,000 nights of concurrent reference and watch sensor data.

9. How the Ground Truth for the Training Set Was Established

The ground truth for the training set was established using concurrent in-lab polysomnography (PSG) and Home Sleep Apnea Test (HSAT) reference recordings. These are the gold standard diagnostic tests for sleep apnea, providing objective measures like the Apnea-Hypopnea Index (AHI). The document also mentions that the distribution of sleep apnea classifications (normal, mild, moderate, severe) was broad in this dataset.

Ask a specific question about this device

The Sleep Apnea Feature is an over-the-counter (OTC) software-only, mobile medical application operating on a compatible Samsung Galaxy Watch and Phone.

This feature is intended to detect signs of moderate to severe obstructive sleep apnea in the form of significant breathing disruptions in adult users 22 years and older, over a twonight monitoring period. It is intended for on demand use.

This feature is not intended for users who have previously been diagnosed with sleep apnea. Users should not use this feature to replace traditional methods of diagnosis and treatment by a qualified clinician. The data provided by this device is also not intended to assist clinicians in diagnosing sleep disorders.

The Samsung Sleep Apnea Feature leverages wrist-worn PPG and actigraphy technology to create an over-the-counter (OTC) assessment of moderate-to-severe obstructive sleep annea for adults. When enabled, the device utilizes the wearable platform's PPG-derived SpO2 to monitor the user's sleep for repetitive, relative decreases in their blood oxygenation indicative of significant breathing disruptions associated with sleep apnea. Each on-demand assessment period requires two successful nights of observation within 10 days. After two qualifying assessment nights. the device will display the result on the wearable, after which, the user is guided to the phone for additional information. This provides the user with health information so that they may seek out medical attention. No raw signal data, including the SpO2 signal, is provided to the user nor is it able to be shared with clinicians.

The Samsung Sleep Apnea Feature consists of two mobile medical applications, one on the wearable (e.g., Samsung Galaxy Watch) and the other on the connected mobile phone (e.g., Samsung Galaxy Phone), both commercial off-the-shelf general computing platforms. Communications between the two devices are accomplished by encrypted Bluetooth/BLE connection via standard protocols for data transfer. The wearable component of the Sleep Apnea Feature runs in the wearable's operating system allowing it to verify the identification/qualification of the hardware, request SpO2/accelerometer signals via private APIs, display information on the screen display, and send data and receive commands to the phone Sleep Apnea Feature on the associated phone. The phone component of the Sleep Apnea feature provides a UI for onboarding, labeling, and status as well as the ability for device updates.

Here's a breakdown of the acceptance criteria and the study proving the device meets them, based on the provided text:

Acceptance Criteria and Device Performance

Study Details

1. Sample size used for the test set and the data provenance:

   • Total enrolled subjects: 620
   • Subjects who completed the investigation: 573 (620 - 47 did not complete)
   • Test set size for sensitivity/specificity calculation: 202 (True Positives + False Negatives) for sensitivity; 268 (True Negatives + False Positives) for specificity. The total for these calculations is 470 subjects.
   • Data Provenance:
     • Country of Origin: Not explicitly stated, but the sponsor information lists "Samsung Research America, Mountain View, CA 94043 USA," suggesting the study was conducted within the United States.
     • Retrospective or Prospective: Prospective, as it states "A non-randomized, open-label, multi-center, single-blind study was conducted in an enriched adult population with an accredited sleep lab recruiting and enrolling subjects..."

2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

   • The ground truth was established by "physician's assessment on corresponding PSG from an FDA-cleared PSG device as the clinical gold standard."
   • The text does not specify the number of physicians or their individual qualifications (e.g., years of experience, specific board certifications), only that it was a "physician's assessment." The study was conducted in an "accredited sleep lab," implying appropriately qualified professionals.

3. Adjudication method for the test set:

   • The text does not explicitly state a formal adjudication method like "2+1" or "3+1" for reading the PSG results to establish ground truth. It refers to "physician's assessment," which typically implies a clinical standard interpretation, likely by qualified sleep physicians within the accredited labs.

4. If a multi-reader multi-case (MRMC) comparative effectiveness study was done, if so, what was the effect size of how much human readers improve with AI vs without AI assistance:

   • No, a multi-reader multi-case (MRMC) comparative effectiveness study comparing human readers with and without AI assistance was not conducted or reported. This study assessed the standalone performance of the AI device against a physician-interpreted PSG gold standard.

5. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:

   • Yes, a standalone performance study was done. The study "compared the result from the Samsung Sleep Apnea Feature, the Device Under Test (DUT), with physician's assessment on corresponding PSG from an FDA-cleared PSG device as the clinical gold standard." This indicates the device's algorithm generated results independently, which were then compared to the ground truth.

6. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):

   • The primary ground truth used was physician's assessment on corresponding Polysomnography (PSG) results from an FDA-cleared PSG device, referred to as the "clinical gold standard." This falls under the category of "expert assessment" based on comprehensive physiological data.

7. The sample size for the training set:

   • The development of the machine-learned algorithms utilized datasets from "over 1000 subjects, split into separate training, tuning, and testing datasets." The exact sample size specifically for the training set is not provided, only the total number of subjects for development phases.

8. How the ground truth for the training set was established:

   • The text states that "datasets from representative populations were utilized from over 1000 subjects" for development, and these datasets were "maintained independently from the final verification and validation activities." It is implied that the ground truth for these development datasets would have also been established using PSG or similar gold standard methods, consistent with clinical practice for sleep apnea diagnosis, but the precise methodology for establishing ground truth specifically for the training set is not detailed.

Ask a specific question about this device