Session Item

Density interfaces that can be found  in a radiotherapy treatment (from air in thoracic cavity to titanium in a prosthesis) make it necessary to pay special attention to the characterization of the different tissues in the planning procedure.



The usual way of stablishing the relationship between the Electron Density of the tissues and the Hounsfield Number (from a CT) in order to calculate a dose distribution as realistic as possible is with a CT2ED calibration curve.



The purpose of this work is to study the performance of the heterogeneity corrections in the Monte Carlo algorithm of Monaco TPS (Elekta) for 6, 10, 6FFF and 10FFF MV photon energies.

CIRS Electron Density Phantom (062M model) is used for two purposes: obtaining CT2ED curves and direct dose measurement with ionization chamber (IC). This phantom is composed of a series of inserts emulating different materials of known densities up to titanium. From a CT image, we can obtain the HU of each material, relate it to its electron densities and set up the curves.

The analysis is done directly comparing the dose calculated with that measured with IC at the same point. A Semiflex chamber is placed in a dedicated insert of the phantom behind the titanium one, and the dose is measured for a 3x3 cm² field and 100 MU. Same configuration was used in Monaco TPS in order to calculate the average dose in the chamber volume.

Three different cases have been studied for dose calculation. First, a curve without titanium is introduced in the TPS (Figure 1). In this case the TPS will use the highest density entered in the curve as the titanium´s. Second, a CT image with titanium is obtained and the corresponding HU is used. The HU obtained now are lower than expected due to saturation of HU values in the CT scan. Finally, the electron density of titanium is forced to its actual value in the structure of interest.

Figure 1: CT2ED curves.

The results obtained for the dose measured with the IC and calculated for different energies, are shown in Table 1.

Table 1: Results

In the first case, without introducing titanium, differences are remarkably high, from 7% to 11%. Higher doses than the real ones are obtained as Monaco is underestimating the attenuation in titanium. Secondly, better results are observed, with differences ranging from 1.4% for 6 MV to 6.6% for 10FFF, however Monaco still calculates wrongly because the density it takes is not the actual one. Finally, when the titanium density is forced, considerably good results are obtained (less than 2%), except for 10FFF energy that we get 5%.

Monaco Monte Carlo algorithm calculates correctly with titanium if the correct electron density is forced, except for 10FFF, where a larger difference is seen. Further invertigation must be done about that.

On the other hand, special care must be taken when entering the CT2ED curve in the TPS, since the problem of HU saturation might be found, using an inaccurate density and then obtaining an inaccurate dose calculation.