The initiative aims to support anaesthesiologists in busy operating rooms by reducing cognitive load, alarm fatigue and human error.
NXP Semiconductors and GE HealthCare have announced a collaboration to advance edge AI innovation in acute care, unveiling new anaesthesia and neonatal care concepts at CES 2026. The partnership combines NXP’s expertise in secure, high performance edge processing with GE HealthCare’s deep experience in medical technology, aiming to deliver low latency, actionable intelligence that supports clinicians in high pressure environments.
The companies are focusing on acute care settings such as operating rooms and neonatal intensive care units, where immediate insights and system resilience are critical. By processing data directly on devices, edge AI enables fast, consistent performance while reducing reliance on cloud connectivity. This approach is intended to give clinicians timely, secure access to critical information at the point of care.
Two proof of concept solutions have been developed. The first applies edge AI to anaesthesia delivery, allowing anaesthesiologists to interact with equipment using real time, hands free voice commands. Designed for busy operating rooms, the concept aims to reduce cognitive load, alarm fatigue and the risk of human error while keeping clinicians focused on patient care.
The second concept supports neonatal monitoring through intelligent, live analysis. Using on device agentic AI, the system can detect infant activity such as crying or resting, identify foreign objects in the cot and recognise potentially unsafe positions. All image processing is performed locally, with no data leaving the device, supporting stringent security and privacy requirements.
Both concepts align with GE HealthCare’s Responsible AI principles, emphasising safety, transparency and fairness. Built on NXP application processors with integrated neural processing units and enabled by the NXP eIQ AI Toolkit, the collaboration highlights how secure edge AI can enhance clinical workflows. The concepts will be demonstrated at NXP’s Pavilion at CES 2026.
