Objective: To design, construct and evaluate the performance of an anthropomorphic phantom for dosimetric audits of head and neck IMRT treatments.
Methods: A customised phantom was commercially manufactured using tissue equivalent materials, containing the anatomical structures of airways, skull, mandible, vertebrae, spinal cord, brain, and parotids. The performance of the phantom was evaluated for ten IMRT plans. Radiochromic film dosimetry was conducted using an A3 flat-bed scanner operating in transmission mode (72 dpi – red channel data). Films were averaged over the last three of six scans, background corrected using a pre-scan image and corrected for scanner light non-uniformity. Accuracy of film registration and positioning were assessed using fiducial marker alignments. The uncertainties of phantom set-up and treatment delivery were quantified by distance-to-agreement and gamma analysis.
Results: The reproducibility of ionisation chamber measurements for each plan was less than 1%. For film dosimetry the uncertainties due to discreteization, positioning of film in the phantom and image co-registration were each approximately 0.5 mm. The largest source of uncertainty was due to positional set-up of the phantom between irradiations. Using gamma analysis criteria of 5%/3mm with a 10% dose threshold, a 90% agreement between the planned and measured relative dose distribution could be readily achieved for all plans.
Conclusion: A dosimetric methodology has been developed using a custom designed anthropomorphic to facilitate the accurate measurement of absolute dose and relative dose distributions for head-and-neck IMRT treatments.