Aidoc and NVIDIA have developed a new open-source framework designed to help chief information officers and governance leaders at health systems manage the fragmentation – across vendors, evaluation processes and IT strategies – that invariably arose with the emergence of artificial intelligence for clinical uses.
The new framework, called BRIDGE, could help standardize validation, interoperability, scalability, deployment and continuous monitoring to aid healthcare systems in achieving faster and more successful AI adoption, said Aidoc in Tuesday’s announcement.
WHY IT MATTERS
BRIDGE stands for Blueprint for Resilient Integration and Deployment of Guided Excellence. The framework aims to help resolve the lack of shared definitions and AI deployment expectations and offers health systems and their vendors a clear, consensus-driven foundation for assessing and integrating machine learning platforms into healthcare delivery.
Built in collaboration with NVIDIA, the framework outlines the technical, regulatory, operational and trust-building criteria that AI tools must meet to be deemed healthcare-ready.
BRIDGE can give health systems the structure they need to implement AI safely and responsibly, according to Dr. Efstathia Andrikopoulou, echocardiography medical director at Harborview Medical Center and associate professor of medicine and collaborative intelligence at the University of Washington.
“Deploying AI at scale requires more than technical performance,” she said in a statement. “It requires trust, transparency and system-level readiness.”
In addition to Andrikopoulou, experts from University Hospitals and Ochsner Health contributed to the roadmap, Aidoc said.
“We’re at a point where AI in healthcare must mature from experimentation to integration,” added Dr. Leonardo Kayat Bittencourt, vice chair of innovation in the University Hospitals Department of Radiology.
The framework guides hospitals navigating the use of clinical AI in differentiating between models and software systems, establishing best practices to ensure a minimum viable production environment, incorporating trust-building mechanisms and scaling systems.
THE LARGER TREND
While AI is already being used more often at the point of care, Aidoc and NVIDIA endeavored in October to create a plan to speed health AI adoption by developing an evidence-based framework.
Aidoc is a vendor of AI tools that integrate real-time insights directly into clinical workflows to help providers close care gaps and accelerate patient access to treatment. NVIDIA offers numerous microservices that can be run from the cloud or on-prem to integrate generative AI into existing applications, focusing on a variety of healthcare use cases, including genomics, imaging and other care delivery priorities, such as predicting hospital readmissions.
By developing BRIDGE in collaboration with providers, academic partners and other industry leaders, Aidoc and NVIDIA said they were striving to build on real-world health AI and focus on common challenges experienced through existing clinical AI integrations.
ON THE RECORD
“To safely deploy AI in healthcare, we need more than strong algorithms,” Reut Yalon, Aidoc’s chief product officer, said in a statement. “We need shared structure…It helps the industry align on what ‘good’ looks like so we can accelerate adoption without compromising safety or performance.”
Kayat Bittencourt added, “BRIDGE gives health systems the foundation they need to scale AI responsibly and the language to do it together.”
Andrea Fox is senior editor of Healthcare IT News.
Email: afox@himss.org
Healthcare IT News is a HIMSS Media publication.
Source link