AI nurse Nurabot steps in as Taiwan leads smart hospital shift

When nurses burn out, robots clock in — and Taiwan’s hospitals are showing the world how it’s done.

Amid a grim World Health Organization warning of a 4.5 million nurse shortage by 2030 due to burnout, Foxconn and NVIDIA are wiring up Taiwan’s hospitals with AI-powered robots.

Meet Nurabot — a collaborative nursing robot built to combat burnout by taking over some of the most taxing tasks in clinical care.

Nurabot isn’t alone. Foxconn has built a suite of smart hospital tools using NVIDIA tech, including AI models that track patient vitals and digital twins that help hospitals design better spaces.

Digital blueprints, real care

Together, these tools aim to transform traditional hospitals into AI-powered ones.

The process starts in the data center. There, giant AI models are trained on NVIDIA supercomputers. Hospitals then test and train robots in virtual twins. Finally, edge devices bring these systems to life in real-time.

Leading medical centers in Taiwan — Taichung Veterans General Hospital, Baishatun Tung Hospital, Mazu Hospital, and Cardinal Tien Hospital — are already on board.

“Taiwan has a highly developed healthcare infrastructure with a strong push toward digital health transformation, creating the ideal environment for robotic integration,” said Shu-Fang Liu, deputy director of the nursing department at TCVGH, which is currently conducting a field trial with Nurabot, in a release.

“Robots are augmenting our capabilities so we can provide more focused, meaningful care.”

This workflow, anchored by NVIDIA technologies such as Jetson Orin, Holoscan, and the Omniverse platform, is already reshaping how hospitals operate, diagnose, and plan.

The company’s AI innovations include FoxBrain, a large language model built with NVIDIA NeMo, and CoDoctor AI, a clinical platform that uses healthcare-specific models for vital sign monitoring, cancer screening, and more.

Foxconn is also contributing to the open-source MONAI ecosystem with CoroSegmentater, a coronary artery segmentation tool designed for improved diagnostic accuracy and preoperative planning.

Less fatigue, more focus

Hospitals are even going a step further. They’re building digital twins of entire wards. These simulations help them test layouts, train robots, and optimize how care is delivered.

TCVGH has already built a twin of one of its nursing stations. Nurabot used it to train virtually before hitting the hospital floor.

The robot itself is a joint project between Foxconn and Kawasaki Heavy Industries. It runs FoxBrain, uses Isaac for Healthcare for virtual training, and carries NVIDIA Holoscan and Jetson Orin onboard for real-time sensing.

In daily use, it handles medicine deliveries, ward patrols, and visitor guidance. Foxconn says it can cut nurse workloads by 30 percent.

“In one of our wards, we are using Nurabot to deliver wound care kits and health education materials to patient bedsides,” Liu added.

“For nurses, having a robot assistant reduces physical fatigue, saving them multiple trips to supply rooms and allowing them to focus more on patients.”

It also helps during visiting hours and night shifts, easing the load when staff is thin. Nurses hope future versions can talk to patients in multiple languages, recognize faces, and even help lift patients.

For example, a lung patient might need two nurses just to sit up and do breathing exercises. With Nurabot lending a hand, one nurse can manage, freeing the other to care for the rest of the ward.

NVIDIA CEO Jensen Huang spotlighted Foxconn’s healthcare push during his COMPUTEX keynote in Taipei. More details are expected at GTC Taipei on May 21–22.