Fig. 2From: External validation: a simulation study to compare cross-validation versus holdout or external testing to assess the performance of clinical prediction models using PET data from DLBCL patientsAUCs of 25 repeats of external simulated datasets with different distributions of Ann Arbor stage in blue with the simulated true expected AUC in black A simulated dataset with 33% Stage 2, 33% Stage 3 and 34% stage 4 patients, B simulated dataset with Stage 2 patients, C simulated dataset with Stage 3 patients, D Simulated dataset with stage 4 patientsBack to article page