Degree Name

Master of Engineering - Research


Faculty of Engineering


Intensity modulated radiation therapy (IMRT) has been widely accepted in clinical applications. Compared to traditional 3D conformal radiation therapy (3D-CRT) techniques, IMRT can deliver the prescribed dose to the target while reducing the dose delivered to surrounding normal tissue. An IMRT treatment composes multiple field segments (typically 100 to 120); these segments usually involve the use of very small fields measuring as little as 1cm x 1cm. Unfortunately, the accuracy of the dosimetry of such small fields has been a concern to medical physicists because different ionisation dosimetry techniques may yield different results mainly due to different detector volumes. This causes a lack of confidence in IMRT quality assurance (QA) results.

This thesis compares and assesses the performance of different dosimeters in small field measurements. The detectors include ionisation chambers of different volumes, a diode, and Gafchromic™ EBT2 film. Tests include verification of the basic properties of the detectors. Their performance is also compared with a radiation therapy planning system (RTPS). The focus of the project is the suitability of these detectors for IMRT QA. This research also develops and validates a simple and novel method of Gafchromic™ EBT2 film analysis for IMRT QA using a grey-scale value to calibrate the film as a viable option compared to the conventional colour channel analysis method.

Results indicate that small-volume detectors have a higher spatial resolution, which is particularly useful in measuring high-dose gradients, such as the buildup region of PDDs or the penumbra of small-field profiles. Among all the detectors the Gafchromic™ EBT2 film has the highest spatial resolution, and its measurement is potentially the closest to the reading from a virtual zero-volume detector. However, these detectors usually have poor precision, as their noise is relatively high, especially the Gafchromic™ film. The poor precision will sometimes reduce the accuracy of the measurement. On the other hand, detectors with larger volumes have worse spatial resolution but a relatively more precise signal. When the detector is used to measure a high dose gradient region, its poor spatial resolution will certainly influence the measurement accuracy. This is usually called the volume-averaging effect. Other factors, such as the detector’s non-tissue equivalence or lack of electron equilibrium in the field, may also affect the accuracy of the measurement.

When scanned as a 16-bit grey-scale value (GSV) image, the Gafchromic™ EBT2 film can provide a very linear GSV-to-dose response in a dose range of 0 to 6Gy. This means that for a relative measurement, conversion to dose using a log or polynomial relation is not required and the GSV pixel value can be linearly related to dose in this dose range. Together with its other advantages, such as extremely high spatial resolution, planar geometry and water equivalence, the Gafchromic™ EBT2 film method is a suitable detector for small-segment IMRT QA. However, due to its poor precision and previously reported batch calibration variation it is not recommended for absolute dose determination. Among the other detectors tested, the CC04 ionisation chamber was most suitable for small-segment IMRT dose determination because it provided the best balance between a reasonably high spatial resolution (to maintain a satisfactory accuracy) and sufficient volume for charge collection (to give relatively high precision).