Quality control of inhomogeneity correction algorithms used in treatment planning systems

This quality control program has been carried out under the auspices of S.F.P.H. (Socíete Fraņcaise des Physiciens d'Hôpital), to evaluate the performances of radiotherapy treatment planning systems (RTPS) used by different institutions. The aim of this Quality Assurance Programme was: (a) to set up a methodology to assess globally the capability of a given system to perform inhomogeneity corrections in the irradiated medium with external photon beams; (b) to analyze the limitations of the algorithms presently used and especially the two-dimensional (2D) dose calculation possibilities; (c) to check, on a number of systems in clinical use, the validity of the method and the variation of the results as compared to measurements used as reference.

Phantom (lung equivalent material placed into polystyrene) measurements, using cobalt-60 radiation, were carried out by the authors. The phantoms were circulated among the participating institutes to be scanned, and used as input to the treatment planning computer.

Ten systems were tested in this study, using seven different inhomogeneity correction algorithms implemented in nine different TPS; four of these algorithms are used in a pixel by pixel basis and five of them in a contour basis. Significant discrepancies or inconsistencies have been observed even for sophisticated models supposed to be mostly accurate.

The proposed tests and the experimental data provided are very useful as part of a quality-control program. They should be included in the initial extensive validation of TPS before starting clinical use, and should be repeated at regular intervals and at each updating of the program. They have the merit of including the whole procedure, from patient data acquisition to dose distribution printout.