Skip to main content

Table 3 Models with different predictors

From: Accuracy of artificial intelligence algorithms in predicting acute respiratory distress syndrome: a systematic review and meta-analysis

Predictors

Study

Sensitivity

Sen-I2

Specificity

Spe-I2

AUC

Images

10

0.81 (0.75–0.86)

93.27 (90.36–96.18)

0.90 (0.82–0.94)

99.25 (99.09–99.41)

0.90 (0.88–0.93)

Images combined with other predictors

8

0.85 (0.80–0.89)

78.94 (64.80–93.08)

0.86 (0.82–0.89)

96.81 (95.57–98.05)

0.92 (0.89–0.94)

Mechanical ventilation parameters

18

0.82 (0.78–0.86)

98.63 (98.37–98.89)

0.78 (0.75–0.81)

98.99 (98.82–99.16)

0.87 (0.83–0.89)

Laboratory assays

31

0.77 (0.70–0.83)

90.26 (87.67–92.86)

0.90 (0.87–0.93)

94.20 (92.88–95.52)

0.91 (0.88–0.93)

Other predictors

14

0.68 (0.62–0.74)

81.65 (72.82–90.48)

0.77 (0.70–0.83)

98.36 (97.98–98.74)

0.78 (0.74–0.81)