Background & objective: CT values are the basis that radiation treatment planning system (TPS) used to calculate the dose distribution of an irradiation treatment plan. However, the CT values of a certain tissue could be very different when it is scanned with dissimilar parameters or acquisition modes, and would add errors to the planning dose calculation. This study was designed to investigate the factors that may impact the calibration of quantitative CT used for radiation treatment planning, and determine the relative importance of these factors.
Methods: CT values of different tissue substitute rods under different CT scanning conditions were measured and compared using RMI 467 phantom. Scanning conditions were classified into 3 modes: (1) evaluating the effect of CT parameters on CT density map by choosing different voltage, slice thickness and acquisition modalities; (2) evaluating the geometric effects of patient size, shape and position in the field of view (FOV) by changing different position of tissue substitute rods in the phantom; (3) evaluating the effect of patient support table on quantitative CT by comparing CT values acquired on patient support table top with those acquired in air. The relative importance leading to CT value error of the 3 modes was also analyzed.
Results: For high-density cortical bone, the applied voltage was the most relevant factor resulting in difference to the reconstructed Hounsfield values of about 150 units, which was 12.7% when scanning voltage changed from 120 KV to 150 KV. The geometric disturbance from different patient sizes and shapes in the FOV could be minimized (less than 10 units for cortical bone). The CT values acquired on the table top varied about 26.34% for water and 293.84% for solid water when compared with those acquired in air.
Conclusions: CT scan voltage and patient support table top scattering make the most impacts to CT values, which may result in error of TPS dose calculation. Inhomogeneous phantoms like RMI 467 can be used to perform QA tests, and calibrations for quantitative CT and reduce the dose error of radiation treatment. It is advisable that a complete set of CT parameters and image acquisition modalities should be determined as part of radiotherapy planning protocols. Patient setup and immobilization modality during CT acquisition should be the same as those during treatment delivery, so as to increase the accuracy of the CT density map for TPS.