Table 4

Summary of performance of outbreak detection algorithms for RRV data from Emerald and Redland.

Method

LGA


Emerald

Redland


TP

FP

FN

(n = 11)

Peak FN

(n = 11)

TP

FP

FN

(n = 11)

Peak FN

(n = 8)


HLM-A

14.3

11.4

60.0

60.0

29.7

2.7

0.0

0.0

(n = 10)

(n = 10)

(n = 9)

(n = 6)

HLM-B

14.7

11.8

60.0

60.0

29.0

7.9

0.0

0.0

(n = 10)

(n = 10)

(n = 9)

(n = 6)

C1-A

7.4

13.0

63.7

63.7

19.0

10.3

45.5

25.0

C1-B

18.2

12.1

54.5

54.5

23.1

19.2

54.5

37.5

C2-A

38.5

30.8

45.5

45.5

33.3

33.3

45.5

25.0

C2-B

50.0

0.0

63.7

63.7

50.0

16.7

54.5

37.5

C2-C

33.3

41.7

63.7

63.7

46.7

33.3

36.4

12.5

C2-D

50.0

0.0

72.7

72.7

44.4

22.2

63.6

50.0

NBC-A

0.0

0.0

100.0

100.0

100.0

0.0

63.6

50.0

NBC-B

100.0

0.0

90.9

90.9

87.5

0.0

36.4

12.5

NBC-C

0.0

0.0

100.0

100.0

100.0

0.0

27.3

0.0

NBC-D

100.0

0.0

90.9

90.9

88.9

0.0

27.3

0.0

NBC-E

100.0

0.0

90.9

90.9

20.0

55.0

63.6

75.0

NBC-F

100.0

0.0

90.9

90.9

23.8

61.9

54.5

62.5

POD-A

50.0

0.0

90.0

90.0

87.5

12.5

33.3

16.7

(n = 10)

(n = 10)

(n = 9)

(n = 6)

POD-B

75.0

0.0

70.0

70.0

71.4

14.3

33.3

16.7

(n = 10)

(n = 10)

(n = 9)

(n = 6)

SaTScan

44.4

55.6

50.0

50.0

85.7

14.3

22.2

0 (n = 6)

(n = 10)

(n = 10)

(n = 9)


Reported values are as per Table 3.

Pelecanos et al. BMC Medical Informatics and Decision Making 2010 10:74   doi:10.1186/1472-6947-10-74

Open Data