Dyania Health, NVIDIA Partner to Bring Clinical Reasoning AI Into Hospital Networks
The collaboration combines Dyania Health’s clinical reasoning technology with NVIDIA’s accelerated computing infrastructure, allowing hospitals to analyze longitudinal patient records across entire populations while keeping patient information behind their own firewalls.
Dyania Health has partnered with NVIDIA to deploy its Synapsis AI clinical reasoning platform within hospital IT networks, enabling health systems to analyze large volumes of unstructured patient records without transferring sensitive data outside their infrastructure.
The collaboration combines Dyania Health’s clinical reasoning technology with NVIDIA’s accelerated computing infrastructure, allowing hospitals to analyze longitudinal patient records across entire populations while keeping patient information behind their own firewalls.
Synapsis AI analyzes medical records, including clinical notes, imaging and operative reports, laboratory results and other unstructured clinical information. The platform answers clinical questions and links its findings to supporting evidence in patient records.
According to Eirini Schlosser, founder and CEO of Dyania Health, the company provides the clinical reasoning technology and implementation expertise, while NVIDIA supplies the computing platform and software required to run the system securely on-premises.
The platform combines specialized AI models, a clinician-annotated clinical knowledge base and evidence tracing. Dyania Health said it has invested more than 25,000 physician hours in annotating medical cases to support its clinical reasoning capabilities.
Synapsis AI is built on NVIDIA’s AI Enterprise software stack. Dyania uses the NVIDIA NeMo framework for model training and reinforcement learning, while NVIDIA TensorRT and TensorRT-LLM support inference within hospital infrastructure.
The companies plan to bring the platform to health systems in the US and internationally, publish joint case studies and integrate Synapsis AI with NVIDIA’s healthcare software stack. They are also exploring technologies to secure AI processing, manage multiple clinical workloads and move from batch processing toward near-real-time clinical insights.
Cleveland Clinic Uses Synapsis AI for Heart Valve Research
Cleveland Clinic used Dyania Health’s Synapsis AI, running on NVIDIA infrastructure, to analyze up to 16 years of medical records from 7,812 patients who underwent aortic valve replacement between 2005 and 2021.
Researchers used the platform to extract information from narrative clinical records and answer 88 questions per patient about their heart valves, medical conditions and subsequent care. Each answer was linked to supporting evidence in the patient's chart.
The resulting dataset was used to develop a predictive model estimating valve deterioration and the likelihood of repeat intervention at three, five and 10 years after the procedure. The estimates can be updated as new patient information becomes available.
"Using technology to bring together and analyze information that can help us predict the duration of implanted heart valves is a promising step towards enhanced outcomes and more personalized care for patients," said Lars G. Svensson, M.D., Ph.D., chief of the Heart, Vascular and Thoracic Institute at Cleveland Clinic, in a statement.
The platform is intended to support clinical research and decision-making, with findings requiring external validation by clinicians.
Stay tuned for more such updates on Digital Health News