Search Results

The Alice 5 System is a Polysomnography System that is intended to record, display and print physiological information to clinicians/physicians. These parameters are presented graphically on a computer screen for diagnostic review, similar in application to the use of a traditional paper based polygraph recorder. The device will be used in hospitals, institutions, sleep centers or clinics, or other test environments where adults or infant patients require the documentation of various sleep or other physiological disorders. This device does not provide alarms and, is not intended for use as an automated apnea monitor.

The Alice 5 is a multi-functional recording device. The system is used for the recording, monitoring, storage and transfer of up to 55 channels of biophysical parameters such as brain, heart and muscle activily, eye movement, blood pressure, breathing, body movements and oximelry. In addition it has 5 auxiliary channels and 2 serial channels for connection of external devices. Electrodes and sensors from the patient are connected to Alice 5 head box which in turn is connected to the Alice 5 base station (recording unit) for data processing and recording. Patient studies recorded using the Alice 5 head box and the base station are viewed, printed, summarized, and analyzed using the Alice Sleepware Soltware.

The provided documentation describes the Alice 5 System, a Polysomnography System, and its acceptance criteria as demonstrated through various performance tests and comparisons to predicate devices.

Here's an analysis of the acceptance criteria and study information:

Acceptance Criteria and Reported Device Performance

Study Details:

1. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective):

   • The document describes bench testing using a "known set of inputs (depicting apneas/hypopneas) was simulated as an input" for the Alice 4R to Alice 5 breath detection comparison. For other comparisons (PTT, Snore, Pressure Based Flow, thermistor channels), the term "assessment" or "comparison" is used for the predicate devices.
   • No specific human subject "test set" sample sizes or patient data provenance (country of origin, retrospective/prospective) are explicitly mentioned for clinical performance evaluation. The comparisons are primarily focused on equivalence of technical signal acquisition and processing with predicate devices through bench testing.
   • The term "performance testing" here primarily refers to engineering and non-clinical bench tests.

2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience):

   • Given the nature of the testing (bench comparisons with simulated inputs or direct signal comparisons with predicate devices), the concept of "experts establishing ground truth for a test set" in the context of clinical performance is not directly applicable.
   • The "ground truth" for the bench tests would have been the "known set of inputs" or the established performance of the predicate device signals as processed and presented.

3. Adjudication method (e.g. 2+1, 3+1, none) for the test set:

   • Not applicable, as the tests described are technical bench comparative assessments rather than clinical performance studies requiring adjudication of human-interpreted results.

4. 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 MRMC comparative effectiveness study was conducted or mentioned. This device is a data acquisition and display system for physiological signals, not an AI-assisted diagnostic or interpretation tool. Its function is to record, display, and print physiological information for clinicians to review, similar to a traditional polygraph, without automated alarms or apnea monitoring.

5. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:

   • The device itself, the Alice 5 System, is essentially a standalone data acquisition system. Its "performance" is based on its ability to accurately record, display, and print physiological data. The bench tests evaluated this standalone performance and its equivalence to predicate devices. There's an "autoscoring algorithm" mentioned for the Alice 4R to Alice 5 comparison, implying some automated analysis capability, but the document doesn't delve into its specific standalone performance metrics or comparative studies for clinical interpretation. The device's indications for use emphasize diagnostic review by clinicians/physicians, implying a human-in-the-loop for interpretation.

6. The type of ground truth used (expert consensus, pathology, outcomes data, etc):

   • The "ground truth" for the bench test comparisons appears to be:
     • Simulated Inputs: For the breath detection (apneas/hypopneas) test, "a known set of inputs" was used.
     • Predicate Device Performance: For other channel comparisons (PTT, Snore, Pressure Based Flow, thermistor), the predicate devices (Alice 4R and Devilbiss Sleep Recorder) served as the reference for equivalent data collection and waveform presentation.
   • There is no mention of ground truth established via expert consensus, pathology, or outcomes data related to clinical diagnosis.

7. The sample size for the training set:

   • Not applicable. This device is not an AI/ML algorithm that requires a "training set" in the traditional sense. It's a hardware system with associated software for data acquisition and display. Software validation (ANSI/AAMI SW68:2001) was performed for the firmware and Alice Sleepware Software, which is more about ensuring proper function and reliability through a defined software life cycle process rather than machine learning training.

8. How the ground truth for the training set was established:

   • Not applicable, as there is no "training set" for this type of device.

Ask a specific question about this device