Skip to main content

Table 4 Performance of predicting ED dispositions by artificial intelligence

From: A meta-analysis of the diagnostic test accuracy of artificial intelligence predicting emergency department dispositions

Metric

Admission

Critical care

Mortality

AUROC

0.866 (0.836–0.929)

0.928 (0.893–0.951)

0.932 (0.894–0.956)

Sensitivity

0.81 (0.74–0.86)

0.86 (0.79–0.91)

0.85 (0.80–0.89)

Specificity

0.87 (0.81–0.91)

0.89 (0.84–0.93)

0.94 (0.90–0.96)

DOR

17.3 (12.40–23.50)

44.5 (24.70–74.20)

74.6 (37.70–133.00)

+LR

4.76 (3.73–6.04)

7.73 (5.06–11.50)

12.8 (7.77–20.10)

-LR

0.277 (0.23–0.33)

0.18 (0.12–0.25)

0.177 (0.13–0.23)

  1. Note: AUROC = Area Under Receiver Operating Characteristic curve, CI = Confidence Interval, DOR = Diagnostic Odds Ratio, ED = Emergency Department, +LR = Positive Likelihood Ratio, and –LR = Negative Likelihood Ratio
Back to article page

BMC Medical Informatics and Decision Making

ISSN: 1472-6947

Contact us