Search Results

The Oxevision Sleep Device is an activity monitor designed and intended for documenting physical movements associated with applications in physiological monitoring. The device's intended use is to analyze subject activity, movement and physiological sign data associated with movement during sleep and to extract information about certain sleep parameters from these movements and physiological sign data.

The device provides a timeline of periods when a bed space is occupied, and periods when a subject is asleep when the bed space is occupied.

The Oxevision Sleep Device is software assessing video from a fixed-installation device for use within single occupancy bed spaces within hospitals, general care and secured environments.

The Oxevision Sleep Device is indicated for use on subjects 18 years of age or older.

Oxevision Sleep is a software-only medical device (SaMD) that provides noncontact sleep assessment in the inpatient setting based on the analysis of patient movement, activity and physiological sign data derived from video, without the need for contact devices to be attached to the patient or bed.

The device consists of custom-designed software assessing video footage collected using off-the-shelf cameras installed within single occupancy bed spaces within hospitals, general care and secured environments. Proprietary software-controlled algorithms are used to derive patient movement, activity and physiological sign data and then to obtain information on bed occupancy and sleep state from the analysis of this data.

The device software automates recognition of sleep periods, generation of sleep reports, and their presentation in a graphical display for use by a healthcare professional.

Here's a summary of the acceptance criteria and the study that proves the device meets them, based on the provided text:

Acceptance Criteria and Device Performance for Oxevision Sleep Device

1. Table of Acceptance Criteria and Reported Device Performance

2. Sample Size and Data Provenance for Test Set

• Sample Size: 60 individuals, resulting in a total of 772.65 hours of data.
• Data Provenance: The text does not explicitly state the country of origin. It mentions "a sample of 60 individuals" and "validation data collected from the 60 adults." The study appears to be prospective as it involved collecting "Reference measurements (physiological signals and video polysomnography data) ... concurrently from a standard off-the-shelf camera and hardware installed in two rooms."

3. Number of Experts and Qualifications for Ground Truth (Test Set)

• Polysomnography (PSG) Scoring:
  • Number of Experts: Three trained sleep physiologists.
  • Qualifications: "trained sleep physiologists, blinded to the video data collected by the standard off-the-shelf camera." They scored in accordance with the American Academy of Sleep Medicine Manual for the Scoring of Sleep and Associated Events version 2.6 of January 2020.
• Bed Occupancy Annotation:
  • Number of Experts: Two reviewers.
  • Qualifications: "blinded to the algorithm development details."

4. Adjudication Method for Test Set

• Sleep State (PSG): The ground truth for sleep state was established using "triple-scored PSG data" with an "epoch-by-epoch majority vote." Epochs where no majority label was available (e.g., due to artifact) were excluded from the analysis.
• Bed Occupancy (Video Annotation): The ground truth for bed occupancy was established by "two reviewers, blinded to the algorithm development details" who "reviewed and annotated" the video data. The specific adjudication method beyond "annotated" by two reviewers is not explicitly detailed (e.g., if discrepancies were resolved by a third reviewer).

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

• The document does not indicate that an MRMC comparative effectiveness study was done to evaluate the effect size of human readers improving with AI vs. without AI assistance. The study focuses on the standalone performance of the algorithm against reference standards.

6. Standalone Performance

• Yes, a standalone (algorithm only without human-in-the-loop performance) study was done. The entire clinical performance section describes the algorithm's performance against established reference standards for bed occupancy detection and sleep/wake classification.

7. Type of Ground Truth Used

• Bed Occupancy: Expert annotation of video data.
• Sleep/Wake Classification: Expert consensus from "triple-scored PSG data" by trained sleep physiologists, adhering to AASM guidelines. This can be categorized as a type of expert consensus based on a gold-standard diagnostic tool (PSG).

8. Sample Size for Training Set

• The document does not explicitly state the sample size for the training set. The "Clinical Performance" section specifically focuses on the "validation data collected from the 60 adults."

9. How Ground Truth for Training Set Was Established

• The document does not explicitly state how the ground truth for the training set (if distinct from the validation set) was established. It only describes the ground truth establishment for the clinical validation test set.

Ask a specific question about this device