A peer-reviewed study published in the New England Journal of Medicine today demonstrates that an AI-powered radiology tool developed by Viz.ai detects early-stage lung cancer on CT scans with 31 percent greater accuracy than experienced radiologists working alone.
The study followed 45,000 patients across 28 hospital systems over two years and found that the AI system identified stage I lung nodules that radiologists missed in 18 percent of cases, while maintaining a false positive rate below 3 percent.
The FDA granted the tool breakthrough device designation, and Medicare has approved a reimbursement code for AI-assisted lung cancer screening effective July 1, 2026, potentially expanding access to AI diagnostics nationwide.
