Figure 3.

Model performance as function of the number of training samples. Average test Dice score ${\mathcal{D}}_A$ on the MPRAGE dataset (N_test = 6) as function of training samples N_train. All models were trained with the same hyperparameters as summarized in Supplementary Table 3. The exact number of training samples for the plotted data points is N_train ∈ {4, 8, 20, 258, 465}.