|| Login  
Krupinski, Johnson, Roehrig, Nafziger, & Lubin (2005)

Reader and case sample combinations needed for 0.80 power based on the data of:

Krupinski EA, Johnson J, Roehrig H, Nafziger J, Lubin J. On-axis and off-axis viewing of images on CRT displays and LCDs: Observer performance and vision model predictions. Academic Radiology 2005;12:957-964.

Readers: 6 board certified family physicians

Cases: 80 breast images (40 with benign and 40 with malignant masses) downloaded from the Digital Database for Screening Mammography (Heath M, Bowyers KW, Kopans D, et al. Current status of the digital database for screening mammography. In: Karssemeijer N, Thijssen M, Hendriks J, van Erning  L, ads. Digital Mammography. Boston, MA: Klu-wer Academic Publishers, 1998; 457-460. (http://marathon.csee.usf.edu/Mammography/Database.htmi. Accessed: June 16, 2005.). A larger series of images was created by extracting 512 X 512 regions of interest (ROIs) around each mass. Digital processing with a median filter was used to remove the masses and generate signal-absent versions of the images. To create lower contrast (75%, 50%, and 25%) versions of the masses, weighted superposition of the signal-present and signal-absent image versions were used. The final test set had five versions of each of the 80 base images (100%, or original contrast; 75%; 50%; 25%; and 0%, or signal-absent), yielding 400 test images. They were down sampled from 512 X 512 to 256 X 256 pixel ROIs.

Conditions: The four viewing conditions were (1) LCD on-axis, (2) LCD off-axis, (3) CRT on-axis, and (4) CRT off-axis.

ROC Model = PropROC, Parameter = AUC,

AUCs = 0.887, 0.910, 0.899, 0.863.

Readers = 6, Cases = 400,

MS( Treatment*Reader ) = 0.05331279,

MS( Treatment*Case ) = 0.10428066,

MS( Treatment*Reader*Case ) = 0.05682980.

 

AUC Difference = 0.03

Readers

Readers

and Cases

Random

Cases

Random

Readers

Random

3

737

471

1114

4

485

388

526

5

385

339

339

6

331

306

250

7

297

282

199

8

273

264

165

9

256

251

141

10

242

239

123

11

232

230

110

12

223

223

99

13

216

217

90

14

210

211

82

15

205

206

76

 

AUC Difference = 0.05

Readers

Readers

and Cases

Random

Cases

Random

Readers

Random

3

288

171

451

4

183

141

200

5

143

123

127

6

122

112

93

7

109

103

73

8

100

97

61

9

94

92

52

10

89

88

45

11

85

85

40

12

81

82

36

13

79

80

33

14

77

78

30

15

75

76

28

 

AUC Difference = 0.10

Readers

Readers

and Cases

Random

Cases

Random

Readers

Random

3

75

45

119

4

47

37

52

5

37

33

32

6

31

30

24

7

28

28

20

8

26

26

20

9

24

25

20

10

23

24

20

11

22

23

20

12

21

22

20

13

20

22

20

14

20

21

20

15

20

21

20

 


   Terms of Use   ||   Privacy Statement   ||   Copyright 2006 Medical Image Perception Laboratory