(401 days)
Onera STS measures and records multiple physiological parameters from a sleep study which are used by clinicians to make a decision on the diagnosis of sleep disorders.
Onera STS is intended to be used on a patient, who has been prescribed a sleep study by a healthcare professional. The device is designed to be used under the direction of a physician or trained technician but applied by a layperson.
The recorded data will be made available to a healthcare professional to assist in the diagnosis of sleep disorders.
The device is intended to be used for adults.
Onera STS is a wearable system for measuring (physiological) signals during a sleep study. The device can be used in home (Home Healthcare Environment) as well as Professional Healthcare Facilities, to perform the sleep study.
Onera STS consists of four sensors applied on the forehead, upper chest area, abdominal and lower leg area.
The sensors measure EEG, EOG, EMG, ECG, bioimpedance based respiratory effort, bioimpedance based respiratory flow, cannula based respiratory flow, oxygen saturation, activity, position, and sound pressure level.
The study preparation and data retrieval are done in a professional/clinical environment by a dedicated trained operator.
The device is not a life supporting physiological monitor
Here's an analysis of the acceptance criteria and the study that proves the device meets the acceptance criteria, based on the provided FDA 510(k) summary for the Onera Sleep Test System (K210593):
1. Table of Acceptance Criteria and Reported Device Performance
The FDA 510(k) summary primarily focuses on demonstrating substantial equivalence to a predicate device rather than explicitly stating pre-defined "acceptance criteria" in a quantitative table format for all parameters. However, we can infer some key performance metrics and their proven results:
| Measured Parameter / Aspect | Acceptance Criteria (Implied/Stated) | Reported Device Performance | Study Type |
|---|---|---|---|
| Biocompatibility | No cytotoxic potential | Exposure of L929 mammalian cell cultures to test item extracts shows no cytotoxic potential. | Lab Testing (Cytotoxicity) |
| Negligible irritant | Electrode and enclosure did not produce any primary dermal reaction after exposure to the skin of New Zealand White Rabbits. | Lab Testing (Irritation or Intracutaneous reactivity) | |
| No sensitization potential | Electrode and enclosure did not induce any skin reaction scores at the challenge exposure following an induction phase when applied topical to albino guinea pigs. | Lab Testing (Sensitization) | |
| SpO2 Measurement Accuracy | ±3% in the range 70-100% (as per ISO 80601-2-61:2019) | Onera STS SpO2 showed an accuracy of ±3% in the range 70-100%. | Clinical Study (SpO2 accuracy validation) |
| Sleep Staging Agreement (Kappa) | Substantial agreement (implied for substantial equivalence) | Cohen's kappa = 0.69 (overall agreement with predicate device). When Stage N1 was removed, kappa reached 0.81. This is generally considered "substantial" to "almost perfect" agreement depending on interpretation. | Clinical Study (Comparative sleep study testing) |
| Sleep Staging Accuracy | High accuracy (implied for substantial equivalence) | Wake: Accuracy 94.08 ± 4.34%, Specificity 97.87 ± 2.00%, Sensitivity 61.92 ± 21.70% | |
| Stage N1: Accuracy 89.62 ± 4.23%, Specificity 95.25 ± 3.02%, Sensitivity 27.19 ± 12.11% | |||
| Stage N2: Accuracy 84.69 ± 4.29%, Specificity 81.55 ± 8.05%, Sensitivity 88.32 ± 4.80% | |||
| Stage N3: Accuracy 95.60 ± 1.80%, Specificity 98.26 ± 2.05%, Sensitivity 76.60 ± 18.58% | |||
| REM: Accuracy 94.70 ± 3.27%, Specificity 95.95 ± 2.66%, Sensitivity 88.12 ± 14.46% | Clinical Study (Comparative sleep study testing) | ||
| Sleep Parameter Correlation | High correlation (implied for substantial equivalence) | Total sleep time: 0.77 | |
| Sleep efficiency: 0.65 | |||
| Sleep latency: 0.95 | |||
| REM onset latency: 0.58 | |||
| Wake after sleep onset: 0.55 | |||
| Minutes in Stage N1: 0.67 | |||
| N2: 0.69 | |||
| N3: 0.65 | |||
| REM: 0.91 | |||
| Wake: 0.64 | Clinical Study (Comparative sleep study testing) |
2. Sample Size Used for the Test Set and Data Provenance
- SpO2 Accuracy Validation:
- Sample Size: 10 healthy volunteers.
- Data Provenance: Not explicitly stated regarding country of origin, but described as a controlled study meeting ISO standards. It is a prospective study as it involved active data collection from subjects for the purpose of validating the device.
- Comparative Sleep Study Testing (Primary Clinical Test Set):
- Sample Size: 32 patients.
- Data Provenance: Not explicitly stated regarding country of origin or if it was retrospective or prospective. However, the description states "concurrently applied to 32 patients undergoing a routine sleep study," which implies a prospective, clinical data collection for the validation purpose.
3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications
- SpO2 Accuracy Validation: The ground truth for SpO2 accuracy is typically established using a reference oximeter that has been calibrated and validated according to specific standards (e.g., arterial blood gas measurements). The document does not specify the number or qualifications of experts involved in the direct "ground truth" establishment, as it's a technical parameter validated against a standard reference.
- Comparative Sleep Study Testing:
- Number of Experts: Not explicitly stated. The document mentions "Studies were scored blinded by a qualified sleep professional." This phrasing suggests at least one, but possibly more, qualified sleep professionals were involved.
- Qualifications: "Qualified sleep professional." Specific details like years of experience or board certification (e.g., diplomate of the American Board of Sleep Medicine) are not provided.
4. Adjudication Method for the Test Set
- SpO2 Accuracy Validation: No explicit adjudication method is mentioned, as SpO2 validation against a reference is a direct measurement comparison.
- Comparative Sleep Study Testing: The scoring was done "blinded" by a qualified sleep professional. It's not explicitly stated if multiple professionals scored and then adjudicated differences (e.g., 2+1, 3+1). The wording "a qualified sleep professional" might suggest single-reader scoring relative to the predicate device, or it might be a simplification of a more complex process. More specific adjudication methods (e.g., consensus, expert panel review) are not described.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study and Effect Size
- No explicit MRMC comparative effectiveness study involving human readers improving with AI vs. without AI assistance was described. This study is focused on the device's ability to record and process physiological parameters similarly to a predicate device, and the sleep staging/scoring is done by a human professional based on the device's output. The device itself is not described as providing AI-assisted interpretations that then improve human reader performance. Its purpose is to provide the raw data for such interpretation.
6. Standalone (Algorithm Only Without Human-in-the-Loop Performance) Study
- The study is essentially a standalone performance evaluation of the device's data collection and preliminary processing capabilities, compared against a predicate device. The "scoring" is done by a human, but the device provides the signals for that scoring. Therefore, the device's performance in generating these signals is evaluated independently of a human's final diagnostic decision. The agreement metrics (Cohen's kappa, accuracy, sensitivity, specificity, correlation) reflect the device's ability to produce data that leads to similar scores as the predicate device.
7. Type of Ground Truth Used
- SpO2 Accuracy Validation: The ground truth for SpO2 was based on a reference standard (likely a co-oximeter or another highly accurate method) as per ISO 80601-2-61:2019, generally derived from arterial blood gas measurements.
- Comparative Sleep Study Testing: The ground truth was established by comparing the Onera STS data to the data from the predicate device (Embletta MPR), with both sets of data scored by a "qualified sleep professional." This means the ground truth for sleep staging and physiological scoring is expert scoring of data from a legally marketed device. It is not stated to be pathology or direct outcomes data, but rather a comparative performance to a widely accepted clinical standard.
8. Sample Size for the Training Set
- The provided document does not mention a specific "training set" or its size for the Onera STS device. This implies that the device's underlying algorithms (if any, separate from standard signal processing) were not developed using a distinct, large, and formally defined training dataset in the way a deep learning AI model would be. The focus is on the device's performance in measuring and recording parameters, and the comparison to an existing predicate device.
9. How the Ground Truth for the Training Set Was Established
- Since a dedicated "training set" is not described, the method for establishing its ground truth is also not applicable or not provided in this document. The device appears to rely on established physiological measurement principles and signal processing, validated against a predicate device and relevant standards.
§ 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).