Fig. 2From: Multiphase CT-based prediction of Child-Pugh classification: a machine learning approachBox-and-whisker plot of the continuous rating scores obtained through linear regression (LR), random forest (RF) and convolution neural network (CNN) for patients of each Child-Pugh class. Thick black horizontal bars, boxes, whiskers, and circles correspond to median, interquartile range, 10th or 90th percentile, and outliers, respectively.Back to article page