GE HealthCare and NVIDIA Have Collaborative Leap Forward for the Development of Autonomous X-Rays

Tweet

GE HealthCare has unveiled a partnership with NVIDIA to develop autonomous X-ray and ultrasound technologies, paving the way for a future where care teams can dedicate more attention to direct patient interaction and complex medical cases.

The healthcare landscape is continuously growing and is driven by innovations, aiming to enhance diagnostic accuracy, improve patient outcomes, and alleviate the burdens faced by clinical staff.
Under the collaboration, GE HealthCare aims to develop AI-enabled X-ray and ultrasound systems through the integration of the new NVIDIA’s Isaac for Healthcare platform, a comprehensive suite of tools designed to facilitate the development of physical AI.

This platform, powered by NVIDIA’s robust computational infrastructure, including Omniverse for robotic simulation, allows GE HealthCare to construct a virtual environment where autonomous ultrasound and X-ray devices can be trained, tested, and refined before deployment.

By using the NVIDIA Cosmos platform for synthetic data generation, physics-based sensor simulation, imitation, and reinforcement learning, GE HealthCare can provide a safe and cost-effective method to expose the algorithms to a wide range of situations that would be ethically challenging to do in a real-world setting.

In the beginning phase, NVIDIA and GE HealthCare will focus on autonomous development within X-ray systems, particularly exploring the potential utilization of the NVIDIA Isaac for Healthcare and Jetson platforms.

In addition, it also uses synthetic data generation to simulate various scenarios, helping to automate repetitive tasks performed by a technologist in the patient exam room. This not only reduces the potential for human error but also frees up valuable time for healthcare professionals to engage in more complex tasks that demand their expertise and critical thinking.

Moreover, the companies are investigating the possibility of developing machine-to-patient interactions, where the system can autonomously guide patients through the entire scanning process, so it can improve patient comfort and compliance.

This new partnership builds upon a pre-existing, long-standing relationship between GE HealthCare and NVIDIA, spanning various areas of GE HealthCare’s business. Under another previous collaboration, GE HealthCare also selected NVIDIA technology to develop its pioneering research foundation model SonoSAMTrack.

SonoSAMTrack is a healthcare-specific research foundation model, trained on approximately 200,000 image-mask pairs. It can deliver impressive accuracy with an average similarity score of over 90%, showing the power of AI in enhancing diagnostic precision today.

Ashton Henning