GE HealthCare (Nasdaq: GEHC) + announced that it utilized Nvidia technology to develop an AI-powered research model called SonoSAMTrack.

SonoSAMTrack combines a promptable foundation model for segmenting objects on ultrasound images called SonoSAM. It segments anatomies, lesions and other essential areas in ultrasound images. GE HealthCare also offers a streamlined version called SonoSAMLite.

The latest development builds on a long-term AI collaboration between GE HealthCare and Nvidia. Nvidia, a leader in AI computing, has worked with a number of medtech companies to incorporate AI into their technologies. Johnson & Johnson MedTech announced this week that it plans to accelerate and scale AI for surgery in partnership with Nvidia.

Asensus Surgical linked up with Nvidia last year to deliver novel clinical intelligence to surgeons in surgical robotics. Medtronic also partnered with Nvidia to enable an AI Access platform to boost the GI Genius intelligent endoscopy module’s capabilities.

“GE HealthCare is committed to investing in innovative technologies that help tackle some of the industry’s biggest challenges. Our vision is to accelerate advancements in medical imaging by introducing foundational AI technologies, thereby empowering data scientists to expedite AI application development and eventually help clinicians and enhance patient care,” said Parminder Bhatia, chief AI officer, GE HealthCare. “By utilizing these versatile, generalist models, we aim to adapt more efficiently to new tasks and medical imaging modalities, often requiring far less labeled data compared to the traditional model retraining approach. This is particularly significant in the healthcare domain, for which data is especially time-consuming and costly to obtain.”

More about GE HealthCare and Nvidia’s plans for foundation models

GE HealthCare says that leveraging AI in healthcare generally requires the retraining of models. This retraining accommodates the unique requirements of different patient populations and hospital settings. That method can lead to heightened costs, complexity and the need for specialized personnel, though.

Foundation models, however, can operate as human-in-the-loop AI systems. GE HealthCare says they can enable swift adaptation to various diseases. That facilitates screening, early detection, tracking progression and identifying non-invasive biomarkers with minimal training requirements.

A recent study showed that SonoSAMTrack demonstrated high performance across seven ultrasound datasets. These encompassed a wide range of anatomies (adult heart and fetal head) and pathologies (breast lesions and musculoskeletal pathologies). They also included different scanning devices.

GE HealthCare said SonoSAMTrack outperformed competing methods “by a substantial margin.”

“Combining Nvidia’s accelerated computing and AI technology stack with GE HealthCare’s medical imaging expertise will help enhance patient care by making ultrasound diagnostics quicker and more accurate,” said David Niewolny, director of business development for healthcare and medical, Nvidia. “This collaboration underscores the importance of using AI for life-saving advancements and setting new standards in healthcare.”