Skip to main content
. 2019 Nov 27;9:1317. doi: 10.3389/fonc.2019.01317

Table 7.

ROC curve results of PSI of each AS types in diagnostic test.

Cancer type Type Cut-off Sensitivity Specificity AUC 95% CI lower 95% CI upper P-value
KIRC
AA 0.616 0.578 0.917 0.715 0.673 0.758 <0.05
AD 0.538 0.653 0.972 0.848 0.813 0.882 <0.05
AP 0.828 0.666 0.903 0.823 0.785 0.862 <0.05
AT 0.862 0.548 0.875 0.720 0.674 0.767 <0.05
ES 0.709 0.707 0.806 0.796 0.755 0.838 <0.05
ME 0.470 0.332 0.875 0.604 0.541 0.667 <0.05
RI 0.587 0.638 0.944 0.814 0.775 0.852 <0.05
Predicted probability 0.908 0.799 0.958 0.935 0.912 0.959 <0.05
KICH
AD 0.540 0.803 0.720 0.779 0.676 0.882 <0.05
AT 0.859 0.864 0.560 0.708 0.565 0.851 <0.05
RI 0.572 0.864 0.640 0.792 0.688 0.896 <0.05
Predicted probability 0.775 0.833 0.960 0.938 0.888 0.988 <0.05
KIRP
AA 0.616 0.645 0.719 0.695 0.612 0.777 <0.05
AD 0.545 0.790 0.688 0.793 0.716 0.869 <0.05
AP 0.827 0.586 0.906 0.770 0.706 0.834 <0.05
AT 0.866 0.817 0.594 0.714 0.607 0.821 <0.05
ES 0.732 0.645 0.781 0.704 0.625 0.784 <0.05
RI 0.598 0.597 0.875 0.787 0.714 0.859 <0.05
Predicted probability 0.855 0.831 0.781 0.875 0.820 0.930 <0.05

AUC, area under the ROC curve; CI, confidence interval.