Table 2 The list of different ML outcomes, along with the pooled estimates for those outcomes that have sufficient records

AUC

32 studies: (Setoyama et al., 2016; Barak-Corren et al., 2017, 2020; Etter et al., 2017; Choi et al., 2018, 2021, 2021; Melhem et al., 2019; McKernan et al., 2019; Cho et al., 2020, 2021; Forkmann et al., 2020; Ge et al., 2020; Machado et al., 2020; Zheng et al., 2020; Van Mensa et al., 2020; Su et al., 2020; Alexopoulos et al., 2021; Bai et al., 2021; Coley et al., 2021; Kim et al., 2021; Adams et al., 2021; Berkelmans et al., 2021; Gradus et al., 2021; Jiang et al., 2021; Harman et al., 2021; Navarro et al., 2021; Balbuena et al., 2022; Beni Edgcomb et al., 2022, 2022; Haroz, E. E. et al., 2022; Miranda et al., 2022)

0.77 (0.74, 0.80)

Accuracy

13 studies: (Delgado-Gomez et al., 2011, 2012; Barros et al., 2017; Bhak et al., 2019; Hill et al., 2019; Cho et al., 2020, 2021; Ge et al., 2020; Zhu et al., 2020; Choi et al., 2021; Kim et al., 2021; Beni Edgcomb et al., 2022; Haroz, E. E. et al., 2022)

0.78 (0.73, 0.84)

Sensitivity

15 studies: (Delgado-Gomez et al., 2012, 2016; Barak-Corren et al., 2017, Barros et al., 2017; DelPozo-Banos et al., 2018; Hill et al., 2019; Melhem et al., 2019; Cho et al., 2020, 2021; Ge et al., 2020; Machado et al., 2020; Choi et al., 2021; Navarro et al., 2021; Beni Edgcomb et al., 2022; Haroz, E. E. et al., 2022)

0.69 (0.60, 0.78)

Specificity

15 studies: (Delgado-Gomez et al., 2012, 2016; Barak-Corren et al., 2017; Barros et al., 2017; DelPozo-Banos et al., 2018; Hill et al., 2019; Melhem et al., 2019; Cho et al., 2020, 2021; Ge et al., 2020; Machado et al., 2020; Choi et al., 2021; Navarro et al., 2021; Beni Edgcomb et al., 2022; Haroz, E. E. et al., 2022)

0.81 (0.77, 0.86)

Positive predictive value

6 studies: (Cho et al., 2020; Machado et al., 2020; Van Mensa et al., 2020; Cho et al., 2021; Navarro et al., 2021; Haroz, E. E. et al., 2022)

0.10 (0.04, 0.19)

Recall

3 studies: (McKernan et al., 2019; Choi et al., 2021; Kim et al., 2021)

0.58 (0.15, 1.29)

Precision

3 studies: (Delgado-Gomez et al., 2016; Choi et al., 2021; Kim et al., 2021)

0.91 (0.85, 0.98)

False negative rate

2 studies: (Cho et al., 2020; Haroz, E. E. et al., 2022)

0.26 (0.24, 0.28)

True positive rate

1 study: (Ballester et al.2021)

0.77 (0.40, 1.34)