Room: Exhibit Hall | Forum 7
Purpose: Commercial CT artifact reduction techniques are offered, but many are impractical and produce inaccurate CT images, especially for dental fillings. This work introduces an innovative artifact management technique that shows dosimetric improvement over commercial metal artifact reduction.
Methods: A CT metal artifact management method for proton planning (AMPP) was created that uses two angled CT scans to generate a single image set with minimal artifacts posterior to the dental metal implants. A dosimetric evaluation was performed using an anthropomorphic phantom with targets and OARs that was modified to allow imaging with and without (baseline) dental metal fillings. An Eclipse proton treatment plan was created and optimized on the phantom baseline scan (no dental fillings). This baseline plan was copied to 3 other CT image sets: uncorrected metal artifact CT, Philips OMAR-corrected CT and our AMPP-corrected CT. Comparisons were performed in a region of interest posterior to the dental fillings.
Results: The targets and OARs were equally identifiable in the AMPP-corrected and baseline CT scans but were obscured in the uncorrected metal artifact CT and OMAR-corrected CT scans. The baseline and AMPP dose distributions were also nearly identical, while the uncorrected CT and OMAR-corrected CT showed a large underdosing of the target and changes to the dose distribution. The baseline, AMPP, metal uncorrected and OMAR-corrected target D95 doses were 60.84Gy, 60.80Gy, 58.08Gy and 58.56Gy, respectively. In addition, the baseline, AMPP, metal uncorrected and OMAR-corrected maximum cord doses were 44.6Gy, 43.0Gy, 48.3Gy and 45.3Gy, respectively. RayStation and Monte Carlo proton dose calculations will also be evaluated using gamma analysis.
Conclusion: The AMPP method provided significantly better visualization of targets and OARs, and a nearly identical dose distribution to the baseline image dose distribution, which was not the case for the uncorrected or OMAR-corrected CT scans.