(70 days)
A wrist-worn activity monitor designed for documenting physical movement associated with applications in physiological monitoring. The device is intended to monitor the activity associated with movement during sleep and make estimates of sleep quantity/quality using accelerometry, based on actigraph algorithms designed specifically for the device's unique signal processing techniques. Can be used to analyze circadian rhythms and assess activity in any instance where quantifiable analysis of physical motion is desirable.
The results of the processed data are graphical and numerical presentations and reports of sleep latency, sleep duration, sleep quality and circadian rhythms for the use by or on the order of physicians, trained technicians, or other healthcare professionals.
The Sleep Watch System is intended for use on a general-purpose computing platform, it does not issue any alarms.
The Sleep Watch System is intended for use in the natural environment for passive, noninvasive, data collection of physiological parameters that will later be transmitted to a SaaS platform for remote review by a clinician. The Sleep Watch device is intended for use in children and older.
The Sleep Watch is a wrist-worn device that monitors activity, temperature, and light exposure, it can be used to analyze sleep quantity and quality, circadian rhythms, automatically collect and store data for sleep parameters, and assess activity, intended for use by or on the order of a Healthcare Professional to aid in the evaluation of sleep disorders based on Actigraphy recordings, typically collected during sleep.
The results of the processed data are graphical and numerical presentations and reports of sleep latency, sleep duration, sleep quality and circadian rhythms for the use by or on the order of physicians, trained technicians, or other healthcare professionals.
The Sleep Watch System is intended for use on a general-purpose computing platform; it does not issue any alarms.
The Sleep Watch system consists of:
- The Sleep Watch built-in with accelerometer, gyroscope, PPG, temperature, and light sensors, as well as BLE and WiFi chips.
- The Sleep Watch collects raw data from each sensor.
- The Sleep Watch processes signals with filters and stores raw data in eMMC storage.
- Psychomotor Vigilance Task (PVT)
- An App manages Sleep Watches
- A web Application Programing Interface (API) to allow authenticated users to upload data collected form Sleep Watch to AMI Cloud Platform
- A database to store the input, intermedium output, final output and associated data.
- A web-based database API to access the database and get outputs.
- A dashboard, a web-based user interface, to display, retrieve, manage, edit, verify, and summarize Sleep Watch outputs.
- Proprietary algorithms to analyze actigraphy.
- A reporting API to generate sleep reports.
The Sleep Watch System is intended for patients in the home environment for passive, noninvasive, data collection of physiological parameters that will later be transmitted to a SaaS platform for remote review by a clinician. The Sleep Watch device is intended for use in children and older.
The Sleep Watch System measures and records:
- PPG (Red, Green, Infrared) raw data
- Accelerometer (X, Y, X) and Gyroscope (Vx, Vy, Vz) raw data
- Light (R, G, B) data
- ZCM (Zero Crossing Mode)
- PIM (Proportional Integrating Measure)
- Estimate Sleep and Wake
- PVT test results
- Skin Temperatures
- MESOR (Midline Estimated Statistic of Rhythm), amplitude, and acrophase
The Sleep Watch allows for on-wrist and/or in-App rating scales (0 to 10), with experimenter selectable initial value (0,5,10) and/or questionnaires (each limited by the constraints of readability). These features should be on-demand, according to an experimenter's selected schedule, or both.
The Sleep Watch device does not provide physiological alarms.
This FDA 510(k) clearance letter and summary for the Sleep Watch device focuses heavily on regulatory compliance, technological comparison, and general software/hardware verification. Crucially, it lacks specific information about clinical performance studies, particularly concerning the quantitative measures of sleep quantity/quality estimates and their accuracy against a gold standard.
Therefore, I cannot fulfill all parts of your request with the provided information. I will construct a response based on the available data, highlighting where information is missing and inferring what would typically be required for such a device clearance.
Here's a breakdown of the acceptance criteria and the study information based on the provided text:
Acceptance Criteria and Device Performance for Sleep Watch
Based on the provided 510(k) summary, the acceptance criteria are not explicitly stated in a quantitative manner (e.g., "accuracy greater than X%"). Instead, the document discusses meeting general design requirements, software verification/validation, and demonstrating substantial equivalence to the predicate device. For a device estimating sleep quantity/quality, performance would typically be assessed by comparing its output to a recognized "gold standard" for sleep measurement, such as Polysomnography (PSG).
Given the absence of specific performance metrics in the provided text, the table below reflects what would typically be expected as acceptance criteria for a device making sleep estimates using actigraphy, and it would normally be accompanied by the device's reported performance against those criteria. As these are not present, I will denote them as "Not Specified in Document."
| Acceptance Criteria Category | Typical Metric (Not Specified in Document) | Reported Device Performance (Not Specified in Document) |
|---|---|---|
| Accuracy of Sleep/Wake Estimation | Sensitivity (true positive rate for sleep) vs. PSG | Not Specified in Document |
| Specificity (true negative rate for wake) vs. PSG | Not Specified in Document | |
| Overall Agreement/Accuracy vs. PSG | Not Specified in Document | |
| Accuracy of Sleep Duration | Mean Absolute Error (MAE) compared to PSG | Not Specified in Document |
| Bland-Altman agreement with PSG | Not Specified in Document | |
| Accuracy of Sleep Latency | Mean Absolute Error (MAE) compared to PSG | Not Specified in Document |
| Reliability/Consistency | Test-retest reliability (e.g., ICC) | Not Specified in Document |
| Usability | User satisfaction, ease of use (qualitative) | "Meets its requirements, performs as intended" (general statement) |
| Safety | Compliance with electrical, biocompatibility, and cybersecurity standards | Compliant to IEC 60601-1, ISO 10993-1, ANSI/UL 2900-2-1, etc. |
| Cybersecurity | Robustness against cyber threats, data integrity | Authentication, authorization, cryptographic controls, etc. |
Study Proving Device Meets Acceptance Criteria
The provided 510(k) summary (Section 7, "Performance Data") describes the testing performed. However, it primarily focuses on non-clinical (software, electrical, and mechanical) testing and verification/validation activities, rather than a clinical performance study demonstrating the accuracy of the sleep estimation algorithms against a gold standard.
Here's the information extracted and inferred from the document:
-
A table of acceptance criteria and the reported device performance:
- As detailed above, specific quantitative acceptance criteria and corresponding reported performance metrics for sleep quantity/quality estimations are not specified in the provided document. The document primarily states that "all pre-defined acceptance criteria for the Sleep Watch were met and all software test cases passed" and that the device "meets its requirements, performs as intended." This refers to internal design and software validation, not clinical performance against a gold standard like PSG.
-
Sample sizes used for the test set and the data provenance:
- Test Set Sample Size: Not Specified. The document refers to "system testing," "verification," and "validation" but does not provide a sample size in terms of patient data or clinical recordings used to validate the accuracy of sleep/wake estimates.
- Data Provenance: Not Specified. There is no mention of the country of origin of any data (clinical or otherwise) or whether it was retrospective or prospective.
-
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Not Applicable/Not Specified. Since a clinical performance study comparing the device's sleep estimations to a ground truth (like PSG scored by experts) is not described in the provided text as part of the "Performance Data," there's no mention of experts establishing ground truth for a test set. This would be a critical component of a clinical validation study for sleep monitoring devices.
-
Adjudication method (e.g., 2+1, 3+1, none) for the test set:
- Not Applicable/Not Specified. As no expert-adjudicated ground truth acquisition process is described for a clinical test set, no adjudication method is 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, not specified. The document does not describe any MRMC study involving human readers or clinicians using or being aided by the Sleep Watch. This type of study would be more relevant to AI-assisted diagnostic tools where human interpretation is central. The Sleep Watch primarily provides processed data and reports for review by clinicians, it's not described as an AI-assistance tool for human interpretation of raw signals.
-
If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- Implicitly, yes, for the algorithm's internal function, but not for its clinical accuracy against a gold standard. The document states "Proprietary algorithms to analyze actigraphy" and "Design validation testing which simulated the intended use to confirm that the end-to-end functionality of the Sleep Watch in conjunction with the actigraphy algorithms meets the design requirements." This suggests standalone testing of the algorithms' functionality. However, it does not confirm a standalone clinical performance study where the device's estimated sleep parameters are compared directly to a clinical gold standard (like PSG) without human intervention in the data acquisition/processing chain beyond collecting the actigraphy data.
-
The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- Not Specified in the context of clinical performance. For the non-clinical testing, requirements were confirmed against "design requirements." For sleep monitoring, the gold standard ground truth would typically be Polysomnography (PSG) data, often scored by certified sleep technologists and overseen by sleep physicians. The document does not state that PSG was used as ground truth for validating the sleep estimation accuracy.
-
The sample size for the training set:
- Not Specified. The document mentions "proprietary algorithms" but does not detail their development, including the size or nature of any training data used for these algorithms.
-
How the ground truth for the training set was established:
- Not Specified. Given the lack of information on training sets, the method for establishing their ground truth is also not mentioned.
Summary of Missing Information Critical for Clinical Performance Evaluation:
The provided 510(k) summary focuses on the technical aspects and regulatory compliance of the Sleep Watch (e.g., software, hardware, safety standards, cybersecurity, and equivalence to a predicate actigraph). It explicitly mentions "Proprietary algorithms to analyze actigraphy" but does not describe the clinical validation study that would typically be performed to demonstrate the accuracy of these algorithms in estimating sleep quantity and quality against a clinical gold standard (like PSG). For a device making sleep estimates, objective clinical performance data (e.g., sensitivity, specificity, accuracy, or agreement metrics against PSG) would be crucial for establishing its effectiveness in its intended use. Without this, the "acceptance criteria" for the clinical performance of its sleep estimation function are not transparent in this document.
§ 882.5050 Biofeedback device.
(a)
Identification. A biofeedback device is an instrument that provides a visual or auditory signal corresponding to the status of one or more of a patient's physiological parameters (e.g., brain alpha wave activity, muscle activity, skin temperature, etc.) so that the patient can control voluntarily these physiological parameters.(b)
Classification. Class II (special controls). The device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter when it is a prescription battery powered device that is indicated for relaxation training and muscle reeducation and prescription use, subject to § 882.9.