(350 days)
The DeVilbiss Sleep Recorder is intended for screening patients suspected of or exhibitina symptoms of sleep disorders. The DeVilbiss Sleep Recorder can be used with an autotitrating CPAP to record the results of CPAP treatment for adults diagnosed with sleep apnea syndrome. Patients suffering from excessive daytime sleepiness should be referred to a sleep disorder specialist. The results of an unattended screening are insufficient to identify all possible medical disorders that may produce these symptoms. This device is intended to aid the physician in diagnosing adult sleep apnea. A qualified medical professional should score the device's recorded signals to determine respiratory events.
The DeVilbiss Sleep Recorder consists of a main control box, a chest mounted interface box and several sensors mounted to the patient. The main control box contains a Novametrix oximeter board, the microcontroller and memory board, four AA battery compartment and connectors for external devices. The main control box is intended to be placed on a night stand or near the patient's bed. A chest mounted interface box containing body position sensors and connections for a nasal thermistor, a nasal cannula, snore microphone and three ECG leads is attached to the patient. The sensors used for a standard diagnostic recording are: 1. ECG leads (3) 2. Snore Microphone 3. Nasal Thermistor or Nasal Cannula 4. Oximeter Probe 5. Body Position sensor (inside chest box)
Here's an analysis of the provided text regarding the DeVilbiss Sleep Recorder (K012437) in relation to acceptance criteria and study details:
1. Table of Acceptance Criteria and Reported Device Performance:
Based on the provided document, explicit quantitative acceptance criteria for device performance are not detailed. The summary states the device's performance tests and clinical trials demonstrated "substantial equivalence to the predicate devices based on types of sensors, battery operation, real time monitoring, presentation and analysis of the recorded data and intended use." It also mentions "tests confirming accuracy of the recorded data to the product specifications and the conformance to electrical standards applied for a Class II Type CF device."
Without specific numerical targets for accuracy, sensitivity, specificity, or agreement with a reference standard, a table of "acceptance criteria" versus "reported performance" cannot be fully constructed.
However, the reported demonstration of performance is that the device achieved substantial equivalence to its predicate devices (Bio-Logic Sleep Scan K962103 and Nellcor Puritan Bennett (Melville) Ltd. Sandman Sleep Data Storage System K934599) through performance tests and clinical trials. This substantial equivalence is the de facto "acceptance" based on the FDA's 510(k) pathway.
2. Sample size used for the test set and the data provenance:
- Sample Size for Test Set: Not explicitly stated. The document refers to "performance tests and clinical trials" without specifying the number of participants or cases involved in these studies.
- Data Provenance: Not explicitly stated (e.g., country of origin). The studies are referred to generally as "performance tests and clinical trials." Given the submission is to the US FDA, it is highly probable the data was generated or is applicable to the US market, but no specific geographic origin is provided.
- Retrospective or Prospective: Not explicitly stated.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- The document states: "A qualified medical professional should score the device's recorded signals to determine respiratory events." and "The information in the study is hand scored for respiratory events and used by a healthcare professional to determine if CPAP treatment is a preferred approach."
- Number of Experts: Not specified.
- Qualifications of Experts: "Qualified medical professional" and "healthcare professional." Specific qualifications (e.g., "radiologist with 10 years of experience" or "board-certified sleep physician") are not provided.
4. Adjudication method (e.g., 2+1, 3+1, none) for the test set:
- Adjudication Method: Not specified. The phrase "hand scored for respiratory events" does not indicate any specific adjudication process if multiple scorers were involved.
5. 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:
- MRMC Study: No. This device is an ambulatory sleep recorder that collects physiological signals. The analysis is "performed on one or more channels to determine SpO2 desaturations, heart rate shifts and Pulse Transit Time (PTT) shifts." There is no mention of "AI assistance" or comparative effectiveness studies involving human readers improving with or without AI. The device functions as a data collection tool, and a human scorer interprets its output.
6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done:
- Standalone Study: The device is not intended for standalone diagnosis. The document explicitly states: "The information in the study is hand scored for respiratory events and used by a healthcare professional to determine if CPAP treatment is a preferred approach." and "A qualified medical professional should score the device's recorded signals to determine respiratory events." This signifies a human-in-the-loop process is mandatory for diagnosis. The device is a "portable recording system" that aids the physician.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- Type of Ground Truth: The ground truth for determining "respiratory events" is effectively expert scoring/interpretation by "a qualified medical professional" or "healthcare professional" based on the recorded signals. This is implicit in the statements regarding hand scoring and professional interpretation. There is no mention of a more objective ground truth like pathology or long-term outcomes data for direct correlation with the device's signal detection.
8. The sample size for the training set:
- Training Set Sample Size: Not applicable/not stated. As described, this device is a physiological signal recorder. There is no mention of a machine learning or AI component that would require a distinct "training set" in the context of supervised learning algorithms. The device's "performance tests and clinical trials" would typically focus on the accuracy of its measurements against established standards or predicate devices, rather than training a diagnostic algorithm.
9. How the ground truth for the training set was established:
- Ground Truth for Training Set: Not applicable, as there's no evident training set for an AI/ML algorithm within the device description. The device's function is to record physiological data accurately. The "ground truth" for its validation would be the accuracy of these recorded signals against known benchmarks or the predicate device's signals, not a diagnostic classification that would typically be trained.
§ 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).