Oxehealth Vital Signs

The Oxehealth Vital Signs system utilizes optical video analysis via fixed cameras and proprietary software algorithms for the non-invasive spot-checking of adult pulse and breathing rates in non-critical healthcare settings. Its methodology relies on algorithmic interpretation of visual cues, primarily chest wall movements, with modifications in this version supporting expanded room coverage through multiple camera setups and an updated software stack including Python.

The Oxehealth Vital Signs device is a software system designed for non-invasive, spot-check measurements of pulse rate and estimated breathing rate derived from chest wall movements. It is intended for use on adult subjects (18 years or older) who are still, located in single occupancy rooms or designated single-occupancy areas within hospitals, general care facilities, and secured environments, specifically for individuals not requiring critical care or continuous vital signs monitoring; operation requires appropriately trained staff with professional healthcare oversight and its use is adjunctive, not the sole method for health assessment. The system processes video footage from fixed-installation, off-the-shelf cameras using custom-designed software with proprietary algorithms (employing C++, Python, and Node.js) to extract these physiological measurements. Performance validation has shown statistical non-inferiority compared to conventional methods, with pulse rate measurements between 50-130 bpm (±3 bpm) and breathing rate measurements between 8-31 breaths per minute (±2 bpm); this modified version maintains core functionality while allowing multiple camera setups to cover separate, non-overlapping areas within a larger room.

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 system's performance vary with patient motion, different skin tones, or ambient lighting changes, and what are the defined operational thresholds for when the "subject is still"?; What specific image processing techniques and algorithms are employed to distinguish true physiological signals (pulse, respiration via chest wall movements) from noise and motion artifacts in diverse environmental conditions?; Considering the use of off-the-shelf cameras and potential deployment in "secured environments," what are the detailed cybersecurity measures, data encryption, and privacy protocols for the video data and derived vital signs?; How does the system manage scenarios where a patient is partially obscured by bedding or changes position frequently, particularly when monitoring separate, non-overlapping areas within a larger room using multiple camera installations?; What are the quantitative performance differences, if any, when using various compatible models of "off-the-shelf" cameras and computing hardware, and how is consistent measurement accuracy ensured across different installations and hardware configurations?