Oxevision Sleep Device

The Oxevision Sleep Device is a software-only medical device intended for non-contact sleep assessment by analyzing patient movement, activity, and physiological sign data derived from video footage in inpatient settings. Its methodology relies on proprietary algorithms to determine bed occupancy and sleep states, with clinical performance validated against polysomnography and human-annotated data to demonstrate its utility for monitoring adults.

The Oxevision Sleep Device is a software-based activity monitor designed for documenting physical movements and associated physiological signs to analyze sleep parameters in adult patients aged 18 and older. Utilizing video footage from fixed-installation, off-the-shelf cameras in single-occupancy bed spaces within hospitals, general care, and secured environments, the device employs proprietary algorithms to assess patient movement, activity, and physiological data. This non-contact system aims to identify periods of bed occupancy and when the subject is asleep, providing a timeline and generating sleep reports for healthcare professionals without requiring any devices to be attached to the patient or bed. The device's clinical performance was validated by comparing its outputs against polysomnography (PSG) as a reference standard and human-annotated bed occupancy data, using a sample of 60 adults. The study demonstrated that the Oxevision Sleep Device achieved high accuracy in determining bed occupancy (over 95%) and strong agreement with PSG for sleep/wake classification (overall agreement >82%, positive agreement >88%, and negative agreement >55%). These results met pre-specified endpoints, indicating the system's capability to reliably extract sleep-related information from video analysis. The software automates the recognition of sleep periods and presents findings in a graphical display for review by healthcare professionals.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. How does the Oxevision Sleep Device's algorithm differentiate between subtle movements indicating restlessness versus actual wakefulness, and how does this compare to traditional polysomnography (PSG) arousals?; What is the range of physiological signs the device can reliably detect and quantify from video footage, beyond general movement, and what is their specific correlation with PSG-derived physiological data?; How robust is the device's performance across diverse patient populations, including those with varying body mass index, skin tones, or underlying conditions that might affect movement patterns or video interpretability?; What are the specific performance metrics for identifying different sleep stages (e.g., REM, NREM stages), if any, beyond just sleep/wake classification?; How does the system handle and report periods of data uncertainty or low confidence, for example, due to occlusions (e.g., blankets over the head) or out-of-view activities, and what is the impact on overall sleep assessment accuracy?