Fig. 2From: A deep-learning approach for segmentation of liver tumors in magnetic resonance imaging using UNet++Scatter plot of actual volume vs. predicted volume of liver tumors in the external test set. The concordance correlation coefficient after the logarithm of the actual and predicted volumes was 0.91, indicating that the volume of the 9 livers in the external test set obtained via automated segmentation was close to the actual volumeBack to article page