Foxconn's Nurabot Cuts Nurse Workload by 30% at Taichung Veterans General Hospital

Foxconn's Nurabot, a collaborative nursing robot built on the NVIDIA Isaac for Healthcare framework, has cut nurse workload by 30% during field testing at Taichung Veterans General Hospital (TCVGH), saving nurses two to three hours per day on transport-related tasks while running navigation tasks at 98% accuracy. "For nurses, having a robot assistant reduces physical fatigue, saving them multiple trips to supply rooms and allowing them to focus more on patients." — Shu-Fang Liu, Deputy Director, Taichung Veterans General Hospital. The deployment is a direct response to a projected global shortfall of 4.5 million nurses by 2030, with Taiwan's hospitals — despite advanced healthcare infrastructure — facing the same burnout pressures from repetitive work such as delivering medication, transporting specimens, and administrative duties that Nurabot is designed to absorb. The robot's software stack layers NVIDIA Isaac Sim with Omniverse for high-fidelity simulation, digital twin creation, and virtual training; Holoscan for real-time sensor data processing; Cosmos for task scheduling and predictive route optimization; and DGX for model training in audio classification and action recognition. Edge inference runs on Jetson AGX Orin modules so the robot can navigate autonomously, deliver medication and specimens, and interact with staff and patients through natural language communication and multimodal perception. Digital-twin validation cut deployment time by 40% compared with a fully physical rollout. Foxconn is rolling the same architecture out to additional medical centers including Cardinal Tien Hospital and Baishatun Tung-Mazu Hospital. TCVGH — Nurabot's first deployment site — appeared on Newsweek's 2025 list of the world's best smart hospitals. Looking ahead, Foxconn plans to extend Nurabot with multilingual communication and patient-mobility support, and is contributing models such as CoroSegmentater for cardiac imaging to the open-source medical community.