Table 1 Summary of quantitative evaluation of liver tumor segmentation. Mean and standard deviation of the different scores for segmentation evaluation are given for algorithms \(A_1\) and \(A_2\). For comparison, the mean Dice score among raters is \(\overline{D(R_i,R_j)} = 0.781 \pm 0.121\). The p-value indicates statistical significance of differences between both algorithms, also after correction for multiple testing.

From: Improving automatic liver tumor segmentation in late-phase MRI using multi-model training and 3D convolutional neural networks