Dyania Health, a New Jersey-based clinical AI company, has developed Synapsis AI, a clinical reasoning engine designed to analyze patient medical records and surface clinical signals for care teams. According to NVIDIA, the system processes free-text notes, operative reports, imaging and pathology reads, and discharge summaries accumulated over years to help identify which patients need attention.
Synapsis AI runs on NVIDIA-accelerated computing inside a health system’s firewall, processing records onsite so patient data remains within the hospital. The company says the tool is assistive: clinicians review the evidence surfaced by the system and make all decisions. Each clinical signal is traceable to its source in the patient record.
Dyania builds and trains its clinical models using NVIDIA NeMo open libraries, then optimizes them with NVIDIA TensorRT and TensorRT-LLM to run on hospital GPU servers, according to NVIDIA.
Cleveland Clinic used Synapsis AI to support a study of more than 7,800 aortic valve replacement patients between 2005 and 2021, with the system analyzing each patient’s record including post-surgical echocardiograms, NVIDIA reports.