RESUMO
PURPOSE: Quantitatively determine an optimum image analysis procedure to mitigate inhomogeneities within the EBT2 film and from scanning for accurate absolute dose measurement deposited by an external radiation therapy beam. Multichannel dosimetry procedures were conceived, described, and quantitatively tested against single and dual channel dosimetry. METHODS: A solid water(TM) block was placed on CT imaging and treatment tables in a configuration that avoids bulky compressive devices. CT markers helped register the CT to the treatment plan and the radiation dose distribution from the radiochromic film. The CT images were digitally rotated and resampled to match the spatial resolution of the scanned dosimetric distribution and treatment plan. The ECLIPSE treatment plan planes were digitally translated through digital triangulation of the treatment isocenter to the CT markers in the CT image. A 6 MV photon beam, conforming to the treatment plan, irradiated the EBT2 film sandwiched between solid water(TM) slabs. The exposed radiochromic film images were rotated and translated to the CT images using coincident markers in the CT image that are associated with "tattoos" marked on the radiochromic film. The exposed radiochromic film gray-levels from a flatbed scanner in reflection mode were converted to dose using calibration films. The test dose distribution was scanned and averaged six times to reduce temporal noise. This study generated dose distributions using the red channel alone, green channel alone, ratio of the red to blue channel, ratio of the green to blue channel, a hybrid approach combining the green to blue ratio for higher doses (>80 cGy) with the red to blue ratio (<80 cGy), multichannel averaging and optimized autonomous multichannel correction. Single channel, multichannel, and channel ratio methods for processing the exposed radiochromic film were compared to the treatment plan via gamma analysis. The ellipsoidal decision surface was defined by its axes of 3% of the maximum dose and 3 mm in the horizontal and vertical directions. RESULTS: The multichannel dosimetry procedures provided excellent agreement with calculation of the dose distribution as determined by the gamma analysis. The green channel mostly performed as well or better than the red channel. The green to blue channel ratio for doses when combined with red to blue ratio ("Hybrid") achieved a high level performance. In addition, new registration procedures were developed and tested for aiding the comparison of calculated and experimentally determined dose distributions. CONCLUSIONS: This study described, developed, and tested new processing methods for reducing inaccuracies in absolute dose determination due to inhomogeneities within the film and from scanning. This study found better performance using optimized multichannel following averaging of all color channels. Combining the channel ratios in a hybrid approach also achieved high performance. Averaging the test films reduced temporal noise that severely degraded the blue channel. This methodology avoided using cumbersome, registered correction matrices. Novel registration and digital rotation of CT images enabled quantitative testing and helped improve contact between the radiochromic film and phantom.