Skip to main content

Table 3 AUROC mean ± std on sepsis and mortality prediction

From: An interpretable risk prediction model for healthcare with pattern attention

  Sepsis prediction Mortality prediction
10 h 8 h 6 h 4 h 10 h 8 h 6 h 4 h
LR .713 ± .017 .739 ± .013 .744 ± .017 .753 ± .018 .803 ± .014 .820 ± .017 .843 ± .014 .866 ± .017
RF .726 ± .014 .741 ± .016 .759 ± .019 .762 ± .016 .813 ± .016 .829 ± .020 .856 ± .016 .873 ± .017
SVM .701 ± .021 .754 ± .019 .761 ± .021 .763 ± .018 .802 ± .021 .835 ± .015 .853 ± .020 .863 ± .018
GRU .748 ± .011 .761 ± .009 .766 ± .008 .767 ± .009 .836 ± .011 .842 ± .009 .861 ± .010 .882 ± .010
LSTM .750 ± .011 .763 ± .010 .767 ± .011 .770 ± .011 .857 ± .009 .862 ± .011 .873 ± .009 .884 ± .011
RETAIN .741 ± .011 .756 ± .009 .758 ± .008 .759 ± .008 .848 ± .007 .860 ± .008 .875 ± .009 .895 ± .008
IFM .746 ± .009 .760 ± .010 .763 ± .009 .765 ± .009 .866 ± .007 .873 ± .008 .882 ± .008 .892 ± .007
\({\text {PAVE}}^{-V}\) .754 ± .006 .768 ± .005 .769 ± .005 .770 ± .006 .886 ± .006 .886 ± .005 .897 ± .005 .915 ± .006
\({\text {PAVE}}^{-T}\) .756 ± .007 .770 ± .006 .770 ± .006 .774 ± .006 .889 ± .008 .892 ± .008 .903 ± .008 .918 ± .007
PAVE .763 ± .005 .773 ± .006 .778 ± .004 .780 ± .004 .897 ± .005 .904 ± .004 .913 ± .005 .922 ± .005