Dosimetry of a low-kV intra-operative X-ray source using basic analytical beam models.

The low energy (30-50 kVp) beams from an intra-operative X-ray source are modelled using a basic analytical model considering just primary beam attenuation and absorption. Spatial dosimetry at such low energies is difficult due to the rapid changes in dose-rate from the radiation source. The purpose of the model was to determine the variation with distance in water of coefficients required for beam dosimetry and to validate beam measurements performed in water of high-gradient dose distributions. The model predicts a change in mean mass-energy absorption coefficient of up to 3 % over the range of clinically-relevant distances in water. Distance-dose distributions (variation in dose with distance in water) for the X-ray source were calculated with the model and found to be in agreement with measurement (at clinically-relevant distances), to within a spatial distance comparable to the dimensions and positional accuracy of the ionization chamber used, and comparable to the expected dosimetric anisotropy of the radiation source. Measured and calculated distance-doses begin to diverge at relatively large distances from the radiation source, which is where dose-rates are so low that detector signal levels are comparable with noise.

The commissioning and acceptance of a commercially manufactured radiotherapy beam tissue compensating filter system.

The principles of tissue compensation in radiotherapy and the methods for their implementation on the GE RT/Plan computer are briefly described. The HEK Medizintechnik GmbH Medical Systems Tissue Compensator cutting device is described, along with the testing procedure adopted for its commissioning and quality assurance testing of the total system. The HEK machine hardware proved to be well engineered, but there were many deficiencies in the software supplied to interface it to the planning computer. Most errors were fatal to the program producing a compensator but a number of errors allowed the cutter to manufacture shapes which on close inspection turned out to be incorrect. These faults were traced to incorrect algorithms and/or incorrect software implementation. It is stressed that the software errors were entirely the responsibility of HEK and that the GE equipment and software were not deficient in any way. However, the experiences of the Adelaide group suggest that caution should be exercised when purchasing two pieces of interacting equipment from separate suppliers.