Figure 1.

Study flow. We extracted the lung adenocarcinoma cases in the cancer genome atlas (TCGA), and classified the patient survival into 4 categories, including alive with no progression, alive with disease, dead with no known disease and dead with disease. We used random forests, support vector machine, and multilayer perceptron to classify the 4-category outcomes. The 5-fold cross-validation approach was used during the tuning and modelling of the machine learning algorithms.