(159 days)
The DormoTech Nlab is a physiological data recorder intended to collect and record data from multiple physiological channels for use by clinical software used in polysomnography and sleep disorder studies. It is intended for use by or on the order of a physician. It is intended for use on patients greater than 6 years of age in a supervised (hospital) or unsupervised (home) environments.
The DormoTech Nlab is a physiological data recorder intended to collect and record data from multiple physiological channels for use by clinical software used in polysomnography and sleep disorder studies. It is intended for use by or on the order of a physician and is intended for use on patients greater than 6 years of age in a supervised (hospital) or unsupervised (home) environment.
It consists of:
The Head Unit
The head unit acquires electric signals indicative of EEG and eye movement located in the upper part of the unit (on the patient's forehead). Plethysmograph for Heart rate and SpO2 measurements, and head relative position to body position are also measured using accelerometer sensors located in the upper part of the unit. The middle part of the unit is located below the mouth, it contains 2 nasal (one in each nostril) and 1 oral airflow sensors (thermistors), 1 EMG sensor, along with a snoring sensor.
The Body Unit
The body unit is made of 2 belts, the upper belt sits on the lower belt sits on the stomach of the patient. Both belts contain respiratory effort and accelerometer sensors, in addition, the upper belt contains an accelerometer to measure body position.
The ExG Unit
The ExG unit is put on using an adhesive sticker on the leg/arm/chest (either leg/arm is ok). It consists of 3 electrodes that acquire leg/arm/chest ExG signal.
The Central Unit
The head, chest and ExG units communicate with the central unit via Bluetooth, the wearable units send the measured data to the central unit receives the data, stores it within an internal flash drive and then transmits the data via Wi-Fi to online servers for further diagnosis. The central unit is located in the test room (up to 10 meters from the patient). The central unit has no contact with the patient.
To ensure device reusability between sessions or patients, sections of the head unit have been designed to be detachable or with barriers between the sensor and patient's skin to stop direct contact. Specifically, the head unit incorporates (1) a detachable nasal and oral airflow section to allow for replacement of the section between each use of the device, (2) a removable barrier over the plethysmograph recorder to create a separation between the patient's skin and the recording device which is replaced between each use of the device and (3) a detachable textile on the inner side of the head unit (i.e., the side in direct contact with the patient's skin) to allow for replacement between each use of the device.
Here's a breakdown of the acceptance criteria and the studies that prove the DormoTech NLab device meets them, based on the provided text:
1. Table of Acceptance Criteria and Reported Device Performance
The acceptance criteria are effectively demonstrated by the "good agreement" between the DormoTech NLab and the gold standard PSG, and the acceptable RMSE for the SpO2 sensor.
| Acceptance Criteria | Reported Device Performance | Comments |
|---|---|---|
| Polysomnography (PSG) Parameters (Compared to Gold Standard PSG) | "Good agreement" between devices, mean difference close to zero, and narrow limits of agreement, indicating interchangeability. Performance considered "substantially equivalent." | |
| AHI (events/h) | Mean Difference: 0.2875 | Accepted |
| ODI (events/h) | Mean Difference: -0.1042 | Accepted |
| Snore (%) | Mean Difference: 1.995 | Accepted |
| Sleep Latency (Minutes) | Mean Difference: 0.9727 | Accepted |
| REM Latency (Minutes) | Mean Difference: -0.2864 | Accepted |
| Wake after Sleep Onset (Minutes) | Mean Difference: -2.091 | Accepted |
| REM (%) | Mean Difference: 0.6045 | Accepted |
| N1 (%) | Mean Difference: -1.659 | Accepted |
| N2 (%) | Mean Difference: -1.095 | Accepted |
| N3 (%) | Mean Difference: 2.173 | Accepted |
| Wake (%) | Mean Difference: -1.127 | Accepted |
| Total Sleep Time (Minutes) | Mean Difference: 4.00 | Accepted |
| Sleep Efficiency (%) | Mean Difference: -0.1773 | Accepted |
| Position (Up) (%) | Mean Difference: -0.2792 | Accepted |
| Position (Supine) (%) | Mean Difference: 1.892 | Accepted |
| Position (Left) (%) | Mean Difference: 0.725 | Accepted |
| Position (Right) (%) | Mean Difference: -0.3042 | Accepted |
| SpO2 Sensor Accuracy (Compared to Arterial HbO2 Saturations) | Root Mean Square Error (RMSE) of 2.53% for 70-100% SpO2, which is explicitly stated as being "within the expected range of accuracy" and allowable per ISO 80601-2-61 (up to 4%). | |
| SpO2 Accuracy (70-100%) | RMSE: 2.53% | Accepted |
| Pulse rate (20-250 bpm) | ±3 bpm | Accepted (similar to reference) |
2. Sample Size Used for the Test Set and Data Provenance
- Polysomnography (PSG) Study:
- Sample Size: 24 subjects (out of 26 recruited) completed the study.
- Data Provenance: Prospective clinical study conducted in two sleep labs in Israel (Shamir Medical Center Be'er Ya'akov, and Millenium Sleep Clinic Be'er Sheva).
- SpO2 Sensor Study:
- Sample Size: 12 patients. 259 data points were included in the analysis.
- Data Provenance: Prospective clinical study. No specific country of origin is mentioned, but it's likely part of the overall clinical trials for the device, potentially in Israel as well.
3. Number of Experts Used to Establish Ground Truth and Qualifications
The document does not explicitly state the number of experts or their specific qualifications (e.g., "radiologist with 10 years of experience") used to establish the ground truth for human-scored parameters in the PSG study. However, the ground truth for the PSG study is based on a "gold standard polysomnogram (PSG) study," which inherently implies scoring by trained sleep specialists.
For the SpO2 sensor study, the ground truth was established by "simultaneous monitoring of arterial HbO2 saturations at six different levels of oxyhemoglobin saturation between 70-100%," which is a direct physiological measurement, not an expert-based ground truth.
4. Adjudication Method for the Test Set
The document does not specify an adjudication method (e.g., 2+1, 3+1) for the human-scored PSG parameters. The mention of "gold standard PSG" suggests that standard clinical scoring practices were followed, which typically involve certified polysomnography technologists scoring sleep studies.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
No multi-reader multi-case (MRMC) comparative effectiveness study was done. The study compares the device's readings directly to a gold-standard PSG for the sleep parameters and directly to arterial blood gas measurements for SpO2, not comparing human readers with and without AI assistance.
6. Standalone (Algorithm Only) Performance
The device is described as a "physiological data recorder intended to collect and record data from multiple physiological channels for use by clinical software used in polysomnography and sleep disorder studies." This implies that the device itself is a data acquisition unit, and its performance is evaluated in terms of its ability to accurately record these physiological signals compared to established methods.
- For PSG parameters: The "DormoTech NLab" is compared to a "gold standard polysomnogram (PSG)." This is a standalone performance evaluation of the device as a data recorder, with the implicit understanding that the data collected by both the NLab and the predicate PSG would then be analyzed by clinical software (or human scorers). The "mean difference" and "limits of agreement" directly assess the NLab's standalone measurement accuracy against the gold standard.
- For SpO2 sensor: The "DormoTech NLab SpO2 sensor" is compared to "arterial HbO2 saturations." This is a standalone performance evaluation of the sensor's accuracy.
7. Type of Ground Truth Used
- Polysomnography (PSG) Parameters: "Gold standard polysomnogram (PSG) study." This refers to a comprehensive sleep study recorded and scored according to established clinical guidelines, serving as the benchmark for various sleep parameters. It's essentially expert-interpreted physiological data.
- SpO2 Sensor: "Arterial HbO2 saturations." This is direct physiological outcomes data obtained through arterial blood gas measurements, which is considered the most accurate measure of blood oxygen saturation.
8. Sample Size for the Training Set
The document does not explicitly state the sample size for any training set. It focuses on the validation studies, which are test sets. This device primarily functions as a physiological data recorder, not necessarily an AI/machine learning algorithm requiring a separate training set in the typical sense. Any internal calibration or algorithm development would likely have used internal datasets, but these are not detailed in this premarket notification.
9. How Ground Truth for the Training Set Was Established
Since no training set details are provided, the method for establishing its ground truth is also not mentioned.
§ 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).