Figure 3

Balanced accuracies obtained for the tumour grade classification task using the 73 textural features only. Bar plots and associated error bars represent the average balanced accuracies and the 95% CIs obtained using all 5 test folds of the cross-validation of the 5 machine learning models as a function of the normalization method and number of bins, respectively. (A) FBN T1w-gd. (B) FBN T2w-flair. (C) FBS T1w-gd. (D) FBS T2w-flair. fixed bin number (relative discretization). FBS fixed bin size (absolute discretization).