AI-Powered Clinical Data Platform Developed for Cardiology Research
Researchers have launched a new clinical data platform that combines approximately 1.18 million ECG records and offers AI-powered querying.
Developed to support cardiovascular research, a unified clinical data platform named VRCardio-Helper-Database integrates electrocardiography records obtained from nine different sources with patient demographic information.
Features and Scope of the Platform
The platform introduced by researchers combines electrocardiography records from nine distinct sources with patient demographic and anthropometric information.
The system contains more than 1,185,800 ECG records and derived signal processing metrics belonging to approximately 432,000 patients.
AI-Powered Query Interface
The platform features an AI-powered interface built on a Model Context Protocol server that translates clinical questions into secure, read-only database queries.
Evaluation and Expert Testing
In evaluations, the query interpretation process achieved a 90 percent precision and 98 percent sensitivity rate. Tested by two senior cardiologists, the majority of free-form research questions were answered successfully.