Artificial Intelligence–Assisted Lung Nodule Evaluation on Low-Dose Chest CT in Asymptomatic Individuals: A Prospective Randomized Controlled Trial

This prospective, single-center, open-label randomized controlled trial (RCT) evaluated the clinical utility of AI-assisted lung nodule evaluation on low-dose chest CT (LDCT) in asymptomatic individuals undergoing routine health checkups. A total of 911 participants were randomized 1:1 to an AI-assisted interpretation group (n=447) or a control group without AI (n=464). In the intervention group, AVIEW LCS (Coreline Soft) automatically detected, classified, and measured lung nodules, with results displayed directly within PACS for review by thoracic radiologists. The primary outcome—interpretation time per examination—showed no significant difference between groups (187 vs 172 seconds; P=.23). However, the AI-assisted group demonstrated significantly higher detection rates of Lung-RADS-positive (category 3 or 4) nodules (16.9% vs 10.3%; P=.03), detection of all nodules (52.9% vs 32.6%; P=.002), and frequency of follow-up LDCT recommendations (15.3% vs 7.4%; P=.04). No lung cancers were diagnosed in either group during the median follow-up of approximately 215–216 days. These findings suggest that integrating AVIEW LCS into real-world PACS workflows can meaningfully improve detection of clinically actionable nodules without increasing interpretation time.