Potential for AI as first reader in lung cancer screening

This study evaluated the potential of using AI as a first reader in lung cancer screening using low-dose CT (LDCT) data from 4,053 participants in the MILD trial. The performance of the Aview LCS AI system was assessed for identifying negative scans. The AI achieved a sensitivity of 88.1% and specificity of 71.4%, with a high negative predictive value of 99.4%. A large proportion of scans classified as negative by the AI were confirmed to be truly negative. These results suggest that AI can effectively function as a first reader, potentially reducing radiologists’ workload by approximately 71% while maintaining safety in screening programs. The study was conducted by researchers from the University of Parma and the Istituto Nazionale dei Tumori using the Aview LCS system.