Fig. 1

Results from the Infarct Evolution Classifier. a Analysis of the Balanced Error Rate (BER) of the independent test set, the correctly classified cases (CCC) and the plot of the three ROC curves of the training set. The best performance was achieved using three features (red arrow). b Image of the voxel position and the mean spectrum ± SD (in gray shading) of the training set for each class with the approximate position of the features selected by the SFFS method indicated by red arrows (see also Table 1). c 2D Fisher’s LDA latent space representing the classification in the training set and the independent test set using three features