ABSTRACT
It should be noted that the systematic errors are the most signicant in terms of undesirable clinical outcomes in EBRT. It is generally believed that the total error in clinical practice of EBRT has a normal distribution and that errors in each step can be added quadratically (van Herk et al. 2000). e total error can then be obtained by adding the standard deviations of each error in quadrature. It is quite obvious that the overall error is dominated by the error in a step with the largest magnitude. erefore, every attempt should be made to reduce that error. Errors cannot be eliminated completely, and a method to accommodate the errors without compromising a positive clinical outcome is to select margins around the clinical target volumes and organs at risk judiciously. It is critically important that the tolerance limits and action levels for the QA test procedures for the dierent steps in the EBRT process preserve the rationale used in the selection of margins for each disease site.
