Skip to main content

Table 2 Screening accuracy based on a per-patient analysis was determined by reading each patient’s in-plane and multiplanar images for 10 seconds and diagnostic accuracy based on a per-rib analysis was determined for reading in-plane images without time constraint

From: Osteoblastic lesion screening with an advanced post-processing package enabling in-plane rib reading in CT-images

 

Ten-second screening accuracy on a per-patient analysis

Diagnostic reading accuracy on a per-rib analysis (no time constraint)

 

In-plane reading

Multiplanar reading

In-plane reading

True positives

23

14

226

True negatives

10

37

993

False positives

27

0

77

False negatives

0

9

5

Sensitivity

100 % (82.1–100 %)

60.9 % (38.8–79.5 %)

97.8 % (94.7–99.2 %)

Specificity

27.0 % (14.4–44.4 %)

100 % (88.3–100 %)

92.8 % (91.0–94.2 %)

Positive predictive value

46.0 % (32.1–60.5 %)

100 % (73.2–100 %)

74.6 % (69.2–79.3 %)

Negative predictive value

100 % (65.5–100 %)

80.4 % (65.6–90.1 %)

99.5 % (98.8–99.8 %)

  1. Multiplanar image analysis of every rib without time constraint was considered to be the reference standard. Confidence intervals are shown in brackets. N = 60 patients/1440 ribs