Clinical AI company Cloudphysician has launched Nightingale, a domain-specific video foundation model purpose-built for continuous, 24/7 visual monitoring of hospital patients. Unlike general-purpose systems, the compact, edge-optimized architecture runs entirely on-premises on a single GPU to process up to 100 live camera feeds, keeping patient data secure within the facility while reducing bandwidth and inference costs by roughly 100x. Trained on thousands of hours of annotated clinical video, Nightingale recognizes subtle physiological signals that periodic physician rounds often miss, such as altered breathing patterns, nasal flaring, tube dislocation, fall risks, and signs of delirium.
Also Read: TransformativeMed and Red Rover Health Partner to Drive Agentic AI Integration in Clinical EHR Workflows
Already deployed across more than 150 hospitals in India, Cloudphysician is now expanding the platform into the U.S. market, targeting the nation’s 1 million hospital beds, of which 200,000 currently feature camera infrastructure projected to reach 500,000 by 2030. Highlight-worthy as an ambient safety layer that integrates directly into existing workflows without altering clinical routines, the platform aims to capture critical changes before a patient’s condition escalates. Explaining the core purpose of the model, Dr. Dhruv Joshi, Co-founder & CEO of Cloudphysician, stated: “Nightingale started in patient rooms, training to pick up the signals a bedside monitor misses and a physician would catch if they could be everywhere at once. This is the kind of AI that can mean the difference between life and death.”
