Relative profile and dose verification of intensity-modulated radiation therapy

Purpose: To develop a quality assurance (QA) procedure to assess the intensity profile and dosimetry for intensity-modulated (IM) treatment fields using electronic portal imaging devices (EPIDs), Methods and Materials: A series of rapidly acquired (similar to 1/sec) portal images are summed and converted to dose, For relative intensity QA, the intended profile is subtracted point-by-point from the measured profile forming a series of error values. The standard deviation, a, of the errors, a measure of the goodness of the match, is minimized by applying a normalization and uniform scatter subtraction from the measured profile. For dose verification (dose to isocenter), an empirically determined phantom-correction factor is added to incorporate the effect of patient presence on EPID readings. Seventy prostate treatment fields were used in a phantom study to verify these approaches, Sensitivity was studied by creating artificial mismatches. Results: The average a for relative profile verification is 3.3% (percentage of average intended intensity) whereas artificial mismatches resulted in a values from 5% to 27%, The average isocentric dose calculated from EPID readings is 1.001 relative to the planned dose with a standard deviation of 0.018. Conclusions: An EPID can be used for profile verification and absolute isocentric dose measurement for IM fields. (C) 2000 Elsevier Science Inc.