ScreenLungNet Outperformed Lung-RADS Assessment Strategy

Background: Lung cancer is the leading cause of cancer and mortality due to cancer worldwide. Early diagnosis has challenges, and global 5-year survival rate is only 10% to 20%. National Lung Screening Trial (NLST) and the NELSON (Nederlands-Leuvens Longkanker Screenings Onderzoek) trial from the Netherlands demonstrated that low-dose CT (LDCT) for lung cancer screening can reduce mortality due to lung cancer by 20% to 24% among individuals who are at high risk of lung cancer. This assessment has been nodule based and may underestimate long-term cancer risk. Other findings, including chronic inflammation, emphysema, pulmonary fibrosis, and global lung abnormalities, may provide additional value for risk prediction. Objective: To assess whether considering additional lung features in addition to nodules could improve long-term lung cancer risk prediction. Design: Retrospective study. Methods: ScreenLungNet is a 3-year lung cancer risk prediction model developed using LDCT data from a more...