Performance of deep learning models for predicting preinvasive, minimally invasive, or invasive adenocarcinoma was improved by combining binary and ternary classification models for predicting invasiveness and adjudication of discordant classification.
The ScreenLungNet model, integrating multiple nodules and global lung features from low-dose CT, outperformed single-feature models in long-term lung cancer risk prediction, and enabled prediction even in screening-negative participants.