Your privacy, your choice

We use essential cookies to make sure the site can function. We also use optional cookies for advertising, personalisation of content, usage analysis, and social media.

By accepting optional cookies, you consent to the processing of your personal data - including transfers to third parties. Some third parties are outside of the European Economic Area, with varying standards of data protection.

See our privacy policy for more information on the use of your personal data.

for further information and to change your choices.
Skip to main content

Table 6 Threshold-specific predictions for discharge, primary, and secondary outcomes per 250 emergency department visits in the CMUH test cohort

From: Using machine learning and natural language processing in triage for prediction of clinical disposition in the emergency department

Threshold

Prediction per 250 ED visits

Discharge

Primary

Secondary

Discharge

Primary

Secondary

True

False

F1

True

False

F1

True

False

F1

0.74a

0.04a

0.22a

145

14

0.849

8

35

0.309

23

25

0.427

0.72

0.06

0.22

149

15

0.857

8

25

0.364

26

27

0.475

0.72

0.08

0.20

149

15

0.857

7

19

0.406

30

30

0.508

0.72

0.10

0.18

149

15

0.857

7

15

0.439

32

32

0.528

0.66

0.12

0.22

157

19

0.875

7

13

0.461

30

24

0.537

0.62

0.14

0.24

162

22

0.883

6

11

0.478

29

20

0.540

0.58

0.14

0.28

166

25

0.889

6

11

0.479

26

16

0.522

0.66

0.16

0.18

158

19

0.875

6

9

0.492

32

26

0.561

0.62

0.18

0.20

162

22

0.883

6

8

0.497

31

21

0.561

0.60

0.20

0.20

165

24

0.887

5

6

0.502

30

20

0.564

  1. CMUH China Medical University Hospital, ED Emergency department
  2. aThe optimal thresholds were determined based on Youden’s J Statistic
Back to article page

BMC Emergency Medicine

ISSN: 1471-227X

Contact us