Table 3 |
||||||||
|
Hypothetical data for two ordered markers, A and B |
||||||||
|
Subjects without cancer (N = 1000) |
Subjects with cancer (N = 100) |
|||||||
|
|
|
|||||||
|
A = 1 |
A = 2 |
A = 3 |
B only |
A = 1 |
A = 2 |
A = 3 |
B only |
|
|
|
||||||||
|
B = 1 |
640 |
80 |
80 |
800 |
20 |
0 |
0 |
20 |
|
B = 2 |
80 |
10a |
10a |
100e |
0 |
20b |
0b |
20f |
|
B = 3 |
80 |
10a |
10a |
100e |
0 |
0b |
60b |
60f |
|
A only |
800 |
100c |
100c |
20 |
20d |
60d |
||
|
|
||||||||
|
For A and B combined; aFalse positive rate (FPR) of indicated region (10+10+10+10)/1000=. 04; bTrue positive rate (TPR) of indicated region = (20 +0+0+60)/100=. 80 For A only; cFalse positive rate (FPR) of indicated region =(100+100)/1000=. 20; dTrue positive rate (TPR) of indicated region = (20+60)/100=. 80 For B only; eFalse positive rate (FPR) of indicated region =(100+100)/1000= .20; fTrue positive rate (TPR) of indicated region = (20+60)/100=.80 |
||||||||
|
Baker et al. BMC Medical Research Methodology 2002 2:4 doi:10.1186/1471-2288-2-4 |
||||||||