(180 days)
ANNE Sleep is a wearable sensor system intended for use in the collection, and storage of physiological parameters to aid in the evaluation of sleep-related breathing disorders of adult patients suspected of sleep apnea. The device is intended for use in the clinical and home setting under the direction and interpretation by a Healthcare Professional (HCP).
ANNE Sleep is a wireless physiological recorder for use as an aid in the diagnosis of sleep-related breathing disorders by healthcare professionals. The system features two skin-mounted, biointegrated sensors that pair with an associated proprietary software application for the transmission and storage of data. The sensors, which are placed on the chest and finger, measure peripheral arterial tonometry (PAT), oximetry (SpQ2), pulse rate, snoring, chest movement, body position, total sleep time, and heart rate. These data are uploaded to a cloud backend that can communicate with a compatible third-party software viewer, where they are reviewed by a qualified healthcare professional to identify sleep-disordered breathing events according to a scoring manual. The number of sleep-disordered breathing events per hour of sleep is the Sibel-AHI, a parameter that functions as an aid to diagnosis of sleep apnea by physicians. A sleep report is generated from this score and uploaded to a web portal, the ANNE Sleep Hub, where it can be stored and further reviewed.
Here's a breakdown of the acceptance criteria and the study details for the ANNE Sleep device, based on the provided text:
Acceptance Criteria and Reported Device Performance
| Acceptance Criteria | Reported Device Performance (ANNE Sleep) |
|---|---|
| SpO2 Accuracy | Average root mean square error (A(rms)) ≤ 3% (range 70-100%) |
| 2.31% (meeting the requirements of ISO 80601-2-61) | |
| Aid to Diagnosis of Moderate to Severe OSA (AHI≥15) | Sensitivity ≥ 90% |
| 90% | |
| Specificity ≥ 98% | |
| 98% | |
| Skin Tolerance (Chest Sensor) | No evidence of increased breakdown or skin dryness; small average increase in temporary redness in 32% of subjects. |
| Electrical Safety & EMC | Compliant with ANSI/AAMI ES60601-1:2005/(R)2012 and IEC 60601-1-2:2014, IEC 60601-1-11:2015. |
| Biocompatibility | Compliant with ISO 10993-1:2018, ISO 10993-5:2009, and ISO 10993-10:2010. |
| Wireless Coexistence | Compliant with ANSI IEEE C63.27-2017. |
| Software Verification & Validation | Compliant with IEC 62304:2015 and FDA guidance. |
| Shelf Life | Demonstrated safe and effective performance over intended device life cycle. |
| Mechanical Durability | Bench tested. |
| Usability | Compliant with FDA guidance. |
| Heart Rate, Body Position, PAT, Pulse Rate, Perfusion Index, Snore, Total Sleep Time, Chest Movement | Performance tested. |
| Precision and Repeatability | Performance tested over multiple nights of sleep. |
| Cybersecurity | Evaluated according to FDA draft guidance. |
| Software of Unknown Provenance | Assessed per FDA guidance. |
Study Details
-
Sample Size used for the test set and the data provenance:
- SpO2 Accuracy Study: n=12 healthy subjects.
- Comparison to PSG Study: n=225 subjects.
- Data Provenance: The SpO2 accuracy study was conducted using healthy subjects. The Comparison to PSG study was a single-arm, open-label, multi-center clinical study conducted in the United States. The study subjects were adults (22 years or older) suspected of having OSA. The demographic breakdown was 44% male, 56% female, and ethnicity was 73% white, 12% Black or African American, 9% Asian, 0.4% American Indian or Alaska Native, and 4% multi-racial. This indicates a prospective clinical study.
-
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- The document does not explicitly state the number of experts used for establishing ground truth for either the SpO2 accuracy study or the PSG comparison study.
- For the SpO2 accuracy, the ground truth was "blood gas analysis", which typically refers to laboratory measurements and not expert interpretation.
- For the PSG comparison, the ground truth was "gold standard, polysomnography (PSG)". While PSG studies are typically scored by trained sleep technologists and interpreted by physicians, the document does not specify the number or qualifications of these individuals involved in establishing the ground truth for this specific test set. The device itself states that its Sibel-AHI is "based on manual scoring of data by HCP," implying that qualified healthcare professionals are involved in interpreting the data for diagnosis.
-
Adjudication method for the test set:
- The document does not specify an explicit adjudication method (e.g., 2+1, 3+1) for either study. For the PSG ground truth, standard PSG scoring guidelines are typically used, which implicitly involves a form of expert consensus, but a formal adjudication process beyond scoring rules is not mentioned.
-
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, an MRMC comparative effectiveness study involving human readers with and without AI assistance was not done or reported. The study compared the ANNE Sleep system's performance as a diagnostic aid (implying human interpretation of the results to aid in diagnosis) against the PSG gold standard. The device generates a "Sibel-AHI (based on manual scoring of data by HCP)," suggesting that human interpretation is an integral part of its intended use but not in an AI-assistance vs. no AI-assistance comparative evaluation.
-
If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- The device's Indications for Use state that it is "intended for use in the clinical and home setting under the direction and interpretation by a Healthcare Professional (HCP)." Furthermore, the "Sibel-AHI" is described as "based on manual scoring of data by HCP." This indicates that the device is not intended for standalone (algorithm-only) performance but rather as an aid requiring human interpretation. The reported sensitivity and specificity relate to the system as a "diagnostic aid," which aligns with the human-in-the-loop model defined in the indications.
-
The type of ground truth used:
- For SpO2 accuracy: Blood gas analysis.
- For OSA diagnosis accuracy: Polysomnography (PSG), considered the gold standard.
-
The sample size for the training set:
- The document does not provide information regarding the sample size used for a training set if machine learning or AI models were used in the device's classification or analysis algorithms. The description focuses on clinical validation, not algorithm development.
-
How the ground truth for the training set was established:
- Since information on a specific training set or machine learning model development is not provided, the method for establishing ground truth for a training set is not available in this document.
§ 868.2375 Breathing frequency monitor.
(a)
Identification. A breathing (ventilatory) frequency monitor is a device intended to measure or monitor a patient's respiratory rate. The device may provide an audible or visible alarm when the respiratory rate, averaged over time, is outside operator settable alarm limits. This device does not include the apnea monitor classified in § 868.2377.(b)
Classification. Class II (performance standards).