Skip to main content

Table 2 Comparison of interexamination variability

From: A comparison of six software packages for evaluation of solid lung nodules using semi-automated volumetry: What is the minimum increase in size to detect growth in repeated CT examinations

Software package
  A B C D E F
a. Common dataset 17.0% 13.1% 20.8% 13.4% 20.5% 19.6%
b. Individual datasets
 Excellent 15.9% 15.4% 21.0% 13.5% 20.4% 19.5%
 Satisfactory 21.9% 18.0% 22.9% 27.6% 19.5% 19.7%
c. Influence of size
 <8 mm 21.2% 18.5% 24.9% 19.5% 25.6% 24.5%
 ≥8 mm 17.1% 14.5% 16.8% 16.1% 12.9% 13.8%
d. Median nodule volume 80 mm3 78 mm3 59 mm3 85 mm3 57 mm3 95 mm3
  1. (a) Variability in the common dataset of 89 nodules that were adequately segmented by all software packages. (b) Variability for the individual datasets of all nodules that were excellently and satisfactorily segmented by each particular software package. A nodule that was rated as excellently segmented in one CT examination and satisfactorily in the other was classified as ‘satisfactory.’ Significant differences are printed bold. (c) Variability for different sizes of nodules, also for the individual datasets. Note that results for different software packages are therefore not comparable. Significant differences are printed in bold. (d) Median nodule volume in the common dataset of 89 nodules that were sufficiently segmented by all software packages. Despite the fact that the sets contain identical nodules, there are substantial differences in nodule volumes measured by the various programs