Fig. 3
![Fig. 3](http://media.springernature.com/full/springer-static/image/art%3A10.1186%2Fs40709-020-00116-3/MediaObjects/40709_2020_116_Fig3_HTML.png)
a The sample distribution ofthe MiR Score in the training set (left), purple and gray bars represent samples of death and survival, respectively; ROC curve analysis of 3- and 5-year survival for MiR Score risk assessment model in the training set (middle); the K–M survival analysis of low- and high- risk groups in the training set (right), HR represents hazard ratio, and the number in parentheses indicates 95% confidence interval (CI). b The sample distribution of the MiR Score in the validation set (left), purple and gray bars represent samples of death and survival, respectively; ROC curve analysis of 3- and 5-year survival for MiR Score risk assessment model in the validation set (middle); the K–M survival analysis of low- and high-risk groups in the validation set (right), HR represents Hazard Ratio, and the number in parentheses indicates 95% confidence interval (CI)