Figure 1.
Anthropomorphic chest phantom with artificial lung nodules at different dose levels.
At the standard CT dose level of 100/120 kVp (a), the 10-mm ground glass nodule (GGN) is easily detectable. At the lowest exposure level of 25 mAs/80 kVp (c), the 10-mm GGN is still visible, but the sensitivity is significantly impaired. The lowest acceptable dose level without a loss of sensitivity for detecting GGNs (12 mm) was 25 mAs/100 kVp (b). For solid nodules, the decrease in detectability was less obvious from the standard dose with a 10-mm nodule (d) to the lowest dose with an 8-mm nodule (f). However, for maintaining sensitivity, a dose of 25 mAs/100 kVp is necessary. (e), Example of a 10-mm nodule. The phantom received only 22% of the standard CT dose (a, d) at the lowest acceptable exposure level (b, e).
Table 1.
Sensitivities of readers and CAD alone and combined for each exposure level.
Table 2.
Interobserver agreement for all nodules.
Figure 2.
Dose dependent sensitivity for solid nodules and subjective image quality.
The sensitivity of radiologists increased between 5 and 11% when CAD was used in combination with human assessment. When the dose was reduced, the subjective image quality dropped faster than the sensitivity, i.e., an accurate diagnosis could be made even when the subjective image quality was reduced.
Table 3.
Mean sensitivities per nodule diameter for each exposure level.
Table 4.
RECIST-diameter- and volume-measuremet-errors of CAD for each nodule size and dose level.