Evidence item

The abstract directly reports both benchmark accuracy figures and identifies the benchmarks as MIMIC-IV-derived. MEASUREMENT

Authenticity UNVERIFIED · extraction UNVERIFIED · independence group none (unreviewed) · created seq 15053

Source: On-premise medical AI agents for reliable clinical decision-making · locator {} · excerpt hash sha256:842bb930a82a696bfe23d755662673e6275f8db3880f93e772b3b8972491fcae

Across two Medical Information Mart for Intensive Care IV (MIMIC-IV)-derived benchmarks, the agent achieved 90.04% accuracy on a seven-disease task and 83.8% accuracy on a four-disease task, approaching a cloud baseline on the primary benchmark.