Measurement error of spiral CT Volumetry: Influence of Low Dose CT Technique

Purpose: To examine the possible measurement errors of lung nodule volumetry at the various scan parameters by using a small nodule phantom. Materials and Methods: We obtained images of a nodule phantom using a spiral CT scanner. The nodule phantom was made of paraffin and urethane and its real volume was known. For the CT scanning experiments, we used three different values for both the pitch of the table feed, i.e. 1:1, 1:15 and 1:2, and the tube current, i.e. 40 mA, 80 mA and 120 mA. All of the images acquired through CT scanning were reconstructed three dimensionally and measured with volumetry software. We tested the correlation between the true volume and the measured volume for each set of parameters using linear regression analysis. Results: For the pitches of table feed of 1:1, 1:1.5 and 1:2, the mean relative errors were 23.3%, 22.8% and 22.6%, respectively. There were perfect correlations among the three sets of measurements (Pearson’s coefficient = 1.000, p