Skip to main content

Table 2 Subgroup analyses of the performance of artificial intelligence in the diagnosis mortality in covid-19 patients

From: The accuracy of artificial intelligence in predicting COVID-19 patient mortality: a systematic review and meta-analysis

Subgroup

Study

Sensitivity

Specificity

PLR

NLR

DOR

AUC

All combined

23

0.82(0.69, 0.91)

0.89(0.79, 0.95)

7.60(4.1, 14.1)

0.20(0.11, 0.35)

38.00(18, 81)

0.92

Model

RF

12

0.78(0.75–0.80)

0.81(0.80–0.82)

5.00(3.41–7.33)

0.21(0.12–0.37)

31.59(13.06–76.40)

0.92

XGBoost

5

0.83(0.79–0.86)

0.87(0.86–0.89)

6.02(3.75–9.65)

0.24(0.14–0.41)

29.82(22.72–39.16)

0.91

LR

6

0.80(0.77–0.84)

0.85(0.84–0.85)

4.32(1.74–10.74)

0.31(0.25–0.38)

12.94(3.52–47.51)

0.86

SVM

4

0.94(0.91–0.97)

0.90(0.88–0.91)

9.48(1.64–57.74)

0.11(0.02–0.47)

90.33(9.84–828.9)

0.98

ANN

4

0.91(0.88–0.94)

0.88(0.88–0.89)

4.85(1.31–17.96)

0.14(0.07–0.26)

43.49(16.88-112.07)

0.94

DNN

3

0.70(0.57–0.82)

0.85(0.81–0.88)

4.12(3.07–5.52)

0.37(0.25–0.55)

11.56(6.11–21.86)

0.83

KNN

3

0.91(0.89–0.93)

0.96(0.95–0.97)

24.63(5.49-110.58)

0.12(0.02–0.85)

231.65(15.92-3369.65)

0.98

GBM

2

0.83(0.76–0.88)

0.72(0.69–0.75)

2.86(2.25–3.65)

0.24(0.17–0.34)

11.98(7.7-18.65)

0.50

DT

2

0.89(0.84–0.92)

0.95(0.94–0.97)

13.17(1.38-125.62)

0.12(0-61.10)

112.39(0.4-31797.54)

0.50

Mortality

0–10%

5

0.67(0.31–0.90)

0.97(0.90–0.99)

19.60(11-34.9)

0.34(0.13–0.91)

58.00(35–95)

0.96

10–20%

8

0.72(0.69–0.75)

0.73(0.72–0.74)

3.62(2.76–4.75)

0.41(0.32–0.53)

8.61(6.46–11.47)

0.80

> 20%

10

0.90(0.85–0.94)

0.87(0.80–0.92)

7.10(4.4–11.3)

0.11(0.07–0.19)

63.00(26–151)

0.95

Center

multicenter

14

0.87(0.86–0.88)

0.89(0.88–0.89)

7.54(5.35–10.62)

0.18(0.12–0.26)

52.58(27.35–101.10)

0.93

monocentric

9

0.77(0.75–0.79)

0.87(0.87–0.88)

4.60(3.18–6.64)

0.33(0.25–0.43)

14.85(9.30–23.70)

0.88

People

Asia

10

0.87(0.86–0.88)

0.95(0.95–0.95)

9.59(5.92–15.55)

0.20(0.13–0.31)

56.94(28.76-112.74)

0.94

Non Asia

13

0.52(0.51–0.54)

0.64(0.63–0.65)

3.09(2.08–4.59)

0.26(0.14–0.48)

12.30(3.83–39.45)

0.84

Outcome

In-hospital mortality

17

0.76(0.75–0.78)

0.85(0.85–0.86)

4.14(3.28–5.24)

0.33(0.26–0.40)

14.50(10.28–20.45)

0.85

  1. Note: LR: logistic regression; SVM: support vector machine; RF: random forest; KNN: K Nearest Neighbors; GBM: Gradient boosting machine; SVM: Support Vector Machine; DNN: Deep Neural Network ; ANN: artificial neural network; XGBoost: eXtreme Gradient Boosting;