Fig. 3
Detection of differences in dose levels used in different hospitals for similar anatomical regions by dose-tracking systems illustrating that different hospitals are using the highest dose depending on the anatomical region involved. a Box plot of the CTDIvol-median value, interquartile range, total range and outliers for all CT-head examinations on the five CT systems with the national P75 and P25 DRLs as benchmark, showing hospital B has the highest CTDIvol. p values (all < 0.01) are the result of comparing the mean rank of each CT system with the mean rank of each other CT system. b Box plot of the CTDIvol-median value, interquartile range, total range and outliers for all CT-pulmonary embolism examinations on four CT systems with the national P75 and P25 DRLs as benchmark, showing the CTDIvol is not statistically different (p value of CT1 versus CT3 ≥ 0.01) between CT system 1 in hospital A and the CT system in hospital B. All other P values (< 0.01) are the result of comparing the mean rank of each CT system with the mean rank of each other CT system. Kruskal-Wallis test with post-hoc Dunn test