Feasibility study of a simple approximation algorithm for in-vivo dose reconstruction by using the transit dose measured using an EPID

The purpose of this study is to verify the accuracy of the dose delivered to the patient during intensity-modulated radiation therapy (IMRT) by using in-vivo dosimetry and to avoid accidental exposure to healthy tissues and organs close to tumors. The in-vivo dose was reconstructed by back projection of the transit dose with a simple approximation that considered only the percent depth dose and inverse square law. While the average gamma index for comparisons of dose distributions between the calculated dose map and the film measurement was less than the one for 96.3% of all pixels with the homogeneous phantom, the passing rate was reduced to 92.8% with the inhomogeneous phantom, suggesting that the reduction was apparently due to the inaccuracy of the reconstruction algorithm for inhomogeneity. The proposed method of calculating the dose inside a phantom was of comparable or better accuracy than the treatment planning system, suggesting that it can be used to verify the accuracy of the dose delivered to the patient during treatment.