Radiation output measurement

12.1. Definitive calibration

 
It is recommended that a definitive calibration (Appendix A) should be performed after the replacement of a major component – such as a monitor ionisation chamber – and at least once a year. The definitive calibration should be carried out accordingly the relevant dosimetry protocol using a dosemeter which has been directly calibrated with a secondary standard chamber. The fact should be taken into account that the dose per pulse varies from one machine to another and this affects the correction of the recombination. If using such a secondary standard dosemeter where polarized voltage can not be changed it is not possible to measure the recombination correction directly and it is therefore necessary to know the dose per pulse. This can be obtained by dividing the dose rate by the pulse repetition frequency (PRF). It is recommended to check it with an oscilloscope because machines sometimes have frequency dividers which make the actual PRF different from the indicated value (e.g. the Elekta SL25 at 25 MV). For the annual calibration it is not necessary to calibrate the dosemeter with another dosemeter, but the measurement should be carried out in a non-standard condition included.

12.2. Routine calibration

 
Ideally the routine calibrations should be carried out following the protocol exactly. However, for convenience it is common practice to use a small water or PMMA phantom. In order to the measurement is classed as a calibration rather than a quick check it should be carried out with a calibrated thimble ionisation chamber and any differences between the dose measured in the calibration phantom and the 30 cm cube water phantom advised in the code of practice should be measured so that the result can be expressed in terms of dose. If a plastic phantom is used for weekly (or daily) checks it may be appropriate to use water on a monthly basis. It is important to record the monitor units indicated by both dosimetry channels. On many accelerators the two channels determine the dose in different areas of the beam. If this is the case a difference between them can be a sensitive indication of a change in beam profile. The machine dose rate indication should also be checked. It is not critical from the point of view of patient dosimetry, but it can be used for an indication of the general health of the machine and if the calibration is inaccurate this can be misleading.

12.3. Constancy checks

 
A number of devices are available which are applicable for a rapid check of dose. These may be parallel plate ionisation chambers with a minimal amount of built-in build-up. This allows measurements to be made at the dose maximum by adding an appropriate thickness of build-up material. In principle, assumed that the build-up is appropriate the calibration factor should be dependent only on the beam quality of the accelerator, but in practice it is found that the calibration may be different for two machines of the same energy. Therefore these devices are inappropriate for calibration purposes, but they are suitable for daily constancy checks. Diodes may also be used for this purpose. Whatever device is used, it should be checked regularly with a calibration measurement. The frequency of such checks depends on the stability of the constancy device. The minimum frequency is 6 months, but every occasion can be appropriate when a calibration is carried out. It is common practice for one dosemeter to be associated with one treatment machine. While ergonomically sound this approach does not benefit from the added security that comes when an instrument is used on several machines, it probably does not develop a change in dosimetry in the same direction at the same time.

12.4. Effect of gantry rotation

 
If the ionisation chamber is damaged it is possible that the plates move as the gantry is rotated. Therefore it is important to check the stability of machine output at different gantry angles. This can be carried out using a thimble ionisation chamber (with an appropriate build-up cap) placed on the axis of the machine. The variation should be compared to the value obtained at the gantry orientation used for calibration. 2 per cent difference is adequate against the absolute calibration. Quick check devices may be susceptible to RF pickup which should be kept in mind as a possible explanation of wrong readings. Where a cobalt-60 unit is available it can provide an excellent method for checking the stability of the constancy meter.

12.5. Wedge factors

 
Wedge factors should be measured by placing the axis of the ionisation chamber parallel to the unwedged direction and making two measurements with the collimator rotated through 180° between them. The mean value must be used to calculate the wedge factor. Measurements should be made at the reference depth (5 or 7 cm). In calculating the wedge factor allowance should be teken for the difference in depth dose with the wedge according to the method adopted by the department. The two measurements should agree to better than 3%. Differences can be associated with the alignment of the beam centre with the wedge or with the accuracy of set -up. The difference in the wedge factor with different orientations of the collimator could cancel out assumed that the wedge factor is stable with orientation of the gantry. Occasionally wedge factors should also be checked with a horizontal beam as the wedge position may have some backlash. Faults in the wedge factor with a horizontal beam may cause a significant error in the dose delivered to the patient. The calibrated dose at any orientation of the collimator or gantry angle should be within 3%. The frequency of wedge factor measurements depends particularly on the design of the accelerator. On some Elekta machines with automatic wedges the wedge factor can be adjusted electronically and therefore it should be checked as frequently as the calibration is carried out. Automatic wedges generally need more frequent measurements than fixed external wedges which can be seen by the assistant. In this case it may be practical to establish a constancy check procedure where the dose is measured with the wedge in the beam for one orientation of the collimator only.

12.6. Output factors

 
The variation of machine output with field size is one of the measurements which should be carried out at commissioning. Although this variation is probably stable over the life of the accelerator it is wise to check a few field sizes on a regular annual basis. This can be most usefully performed by using the machine data as used for routine calculations to calculate the monitor units required for delivering a dose in a few randomly selected depths and field sizes. If differences are experienced further measurements are needed to determine the cause of the discrepancy.

12.7. Depth dose and profiles

 
Depth doses and beam profiles should also be measured at the time of commissioning. Machine characteristics may change over the life of the machine and it is recommended that some of these measurements should be repeated on an annual basis. These measurements also provides added confidence in the original measurements and information about a measure of the stability of the machine performance. The measurements must show that machine dosimetry is consistent within 3%. A discrepancy which is greater with 2% than the value obtained at the commissioning measurements should be investigated.

12.8. Other checks

 
Occasionally the consistency of dose output should be checked. For this it has to take a series of ten consecutive measurements. The standard deviation should be better than 1%. The linearity of the dosemeter should also be checked. It may be experienced that for short exposures the dose delivered is not in proportion to that delivered with longer exposures. It can be caused by instability of the beam before servos can stabilise it or by the errors in the circuits that switch off the beam. A differential diagnosis can be performed by plotting dose delivered against dose set. The possibility of variation of delivered dose with output dose rate should also be taken into account. However, changes in dose rate are usually associated with changes in beam steering, etc. so it is not practical to measure the variation with dose rate routinely. Changing of the PRF does not usually change the dose per pulse and probably it does not affect the dose calibration, for instance there are no any leaky ionisation chamber. The cause of this variability include recombination in the ionisation chamber and changes in beam characteristics with dose rate.

---