On August 27, 2025, Cleveland Clinic and Dyania Health announced a collaboration to integrate Dyania Health’s Synapsis Artificial Intelligence platform across Cleveland Clinic’s clinical research enterprise following pilot programs in cardiology, oncology and neurology. The partnership aims to accelerate clinical trial recruitment by using medically trained large language models to automate chart review, turn unstructured medical records into searchable structured data and scale patient identification for research studies. Cleveland Clinic has also invested in Dyania Health and may benefit financially from the sale of the technology.
Pilot results cited in the announcement highlighted substantial time savings and maintained clinical-grade accuracy. In a melanoma trial, Synapsis Artificial Intelligence identified an appropriate patient in an average of 2.5 minutes with 96 percent accuracy, compared with 95 percent accuracy in 427 minutes by a melanoma-specialized nurse and 88 percent accuracy in 540 minutes by an oncology research nurse. In cardiology, the platform analyzed more than 1.2 million patient records, reviewed 1,476 in one week and correctly identified 30 eligible participants for a Phase 3 transthyretin amyloid cardiomyopathy trial, compared with 14 patients identified over 90 days using routine recruitment. The system also generated understandable justifications for inclusion or exclusion criteria and identified patients across a broader range of clinical sites within the health system.
The organizations said the technology integrates with electronic medical records and abstracts and interprets data such as clinical notes, imaging and pathology, combining those inputs with structured information like organ function and age to match complex, trial-specific eligibility criteria in real time. Teams at Cleveland Clinic and Dyania Health are further validating and deploying applications in neurology, including movement disorders, by annotating de-identified records to benchmark and improve accuracy. Executives from both organizations framed the work as a way to reduce traditional enrollment bottlenecks, expand patient access to trials and build a research infrastructure that complements clinician expertise.
