ROC curve for lung cancer classification. ROC curves for lung cancer classifications (control vs. cancer cases) were constructed for the proposed GDFS method (blue), CFS (green), and rpart (orange). In each case, the posterior probability of belonging to the lung cancer group was calculated for each observation j using features selected, and model parameters estimated, with observation j omitted (leave-one-out cross-validation). ROC curves were constructed from these posterior probabilities using the ROCR package in R . ‘X’ marks the 0.5 classification threshold on each ROC curve.