Feasibility evaluation of kilovoltage cone-beam computed tomography dose calculation following scatter correction: investigations of phantom and representative tumor sites.

Background: To study the feasibility of kilovoltage cone-beam computed tomography (KV-CBCT) dose calculation following scatter correction.

Methods: CIRS 062 and Catphan 504 phantoms were used in this study, and 40 randomly selected subjects representing a variety of cases (ten head cancer cases, ten chest cancer cases, ten abdominal cancer cases and ten pelvic cavity cancer cases) were enrolled. We developed in-house software called the cone-beam CT imaging toolkit (CITK) to improve the quality of CBCT images. We first aligned each planning computed tomography (pCT) image with the corresponding CBCT image using rigid registration after scatter correction. Hounsfield unit-relative electron density (HU-RED) calibration was applied to the CBCT images. The pCT plan was then recalculated on CBCT images. Finally, the dosimetric differences between the two plans were evaluated. The dosimetric parameters included the D98, D2, Dmean, conformity index (CI), homogeneity index (HI) and other organ at risk (OAR) dose parameters of the planning target volume (PTV). The dose distribution index (DDI) and the gamma index were also assessed. Paired Student's t-tests or Wilcoxon rank tests were used to evaluate differences. P<0.05 was considered significant.

Results: In the phantom and patient cases, the average dosimetric difference was less than 1% in the PTV and OARs. There was no significant difference in the CI or HI between the two plans. The gamma pass rate of 2%/2 mm was greater than 95% in both plans. There was a significant difference in the DDI between the two plans in the chest group but not in the other groups.

Conclusions: The results suggest that CBCT has high accuracy in dose calculation via scatter correction and HU-RED calibration.