GGO and minimally invasive adenocarcinomas: how to deal with?

In this issue of the Journal, Mai et al. report on the development of a model that predicts the malignancy and invasiveness of ground-glass nodules (GGNs) based on CT-scan features (1). Similar to previously reported studies, the proposed model showed random forests could achieve 95% and 83% accuracy to predict malignancy and invasiveness, respectively.