Table 1 Individuals’ baseline characteristics of included studies.

From: Diagnostic accuracy of artificial intelligence in detecting left ventricular hypertrophy by electrocardiograph: a systematic review and meta-analysis

Authors

F De la Garza-Salazar11

T Kokubo10

Kwon9

Liu12

Liu16

Sparapani13

Zhao14

Khurshid15

Year

2020

2022

2020

2022

2023

2018

2022

2021

Country

Mexico

Tokyo

Korea

Taiwan

Taiwan

USA

China

England, Wales, Scotland

Study design

Case–control study

Retrospective cohort study

Retrospective cohort study

Retrospective cohort study

Retrospective cohort study

Prospective cohort study

Retrospective cohort study

Prospective cohort study

Population

Patients underwent an echocardiogram and an electrocardiography

Patients underwent electrocardiography within 4 weeks of the transthoracic echocardiography

Patients who underwent electrocardiography and echocardiography within 4 weeks

Patients who underwent electrocardiography and echocardiography within 30 days

Patients underwent an echocardiogram and an electrocardiography

Patients underwent a cardiac magnetic resonance imaging and an electrocardiography

Patients underwent an echocardiogram and an electrocardiography

Patients underwent a cardiac magnetic resonance imaging and an electrocardiography

Diagnostics indicate LVH

Echocardiogram: Two-dimensional linear method

Echocardiogram: Two-dimensional linear method

Echocardiogram: Two-dimensional linear method

Echocardiogram: Two-dimensional linear method

Echocardiogram: Two-dimensional linear method

cMRI

Echocardiogram: Two-dimensional linear method

cMRI

Defined LVH by

Left ventricular mass index (LVMI)

Left ventricular mass index (LVMI)

Left ventricular mass index (LVMI)

Left ventricular mass index (LVMI)

Left ventricular mass index (LVMI)

Left ventricular mass (LVM)

Left ventricular mass index (LVMI)

Left ventricular mass (LVM)

Cut off of LVMI

Male 115 g/m2

Female 95 g/m2

Male 101 g/m2

Female 85 g/m2

Male 132 g/m2

Female 109 g/m2

Male 132 g/m2

Female 109 g/m2

Male 115 g/m2

Female 95 g/m2

The sex-specific 95th percentile of normalized LVM

Male 115 g/m2

Female 95 g/m2

The sex-specific 90th percentile of normalized LVM

AI methods

C5.0 algorithm

ENN

ENN

CNN

BPN

BART

CNN

CNN

Accuracy

0.714

0.732

0.88

0.961

Sensitivity

0.796

0.605

0.454

0.903

0.966

0.29

0.68

0.34

Specificity

0.53

0.813

0.951

0.693

0.956

0.946

0.57

0.96

Subgroup

LVH

Non LVH

LVH

Non LVH

LVH

Non LVH

LVH

Non LVH

LVH

Non LVH

LVH

Non LVH

LVH

Non LVH

LVH

Non LVH

Total number (N)

202

230

1249

6109

4353

16,933

2435

26,310

173

779

69

871

932

931

4777

126

Mean age (years)

69.3 ± 12.1

65.7 ± 14.8

63.31 ± 17.07

67.61 ± 12.46

57.39 ± 14.89

48.6 ± 8.8

44 ± 11.3

61.0 ± 9.9

67.3 ± 10.5

63.9 ± 11.3

63.6 ± 7.7

55.5 ± 14.6

Male (%)

55.6%

16.3%

33.2%

53.6%

52.2%

44.7%

89.9%

46.6%

45.2%

70.7%

47.5%

54%

QUADAS score

12

13

12

13

13

13

12

13

  1. N/A.