Skip to main content

Table 1 Statistics of extracted phrases for all versions

From: Mining of EHR for interface terminology concepts for annotating EHRs of COVID patients

Version

Procedure

Total # of phrases extracted

# phrases accepted after 1st review

% phrases accepted after 1st review

# phrases accepted during 2nd review

% phrases accepted after two reviews

# phrases accepted after F2F meeting

Total # of phrases accepted

% retained w.r.t. total # phrases extracted

CIT_V1.1

Concatenation

1197

251

20.97%

125

31.41%

120

525

43.86%

CIT_V1.2

Anchoring

2488

663

26.65%

198

34.60%

52

937

37.66%

CIT_V2.1

Concatenation

972

449

46.19%

70

53.40%

33

556

57.20%

CIT_V2.2

Anchoring

1345

186

13.83%

41

16.88%

43

280

20.82%

CIT_V3.1

Concatenation

395

82

20.76%

42

31.39%

29

154

38.99%

CIT_V3.2

Anchoring

206

47

22.82%

–

–

18

66

32.04%

CIT_V4.1

Concatenation

101

9

8.91%

–

–

17

26

25.74%

CIT_V4.2

Anchoring

133

8

6.02%

–

–

1

9

6.77%