Skip to main content
Fig. 11 | BMC Bioinformatics

Fig. 11

From: Which data subset should be augmented for deep learning? a simulation study using urothelial cell carcinoma histopathology images

Fig. 11

Testing accuracy and area under receiver operating characteristic curve versus mean training time. Scatter plots between testing accuracy (above) and ROC AUC (below) on one hand and the logarithmic transformation of the mean training time on the other hand, stratified by network. Augment-first models and non-augmented test set results were excluded. Blue solid lines are fitted regression lines. Blue dashed curves are the 95% confidence bands of the forecast (which include the prediction uncertainty of both the mean and the residual). Pearson’s product-moment correlation coefficient (r) and its p value are shown for each plot. ROC AUC = area under the receiver operating characteristic curve

Back to article page