Purpose: The purpose of the present study was to test the reliability in the detection of specific errors introduced randomly in the MLC by the systems used for arc therapy quality assurance. Material and Methods: Intentional errors were introduced in the MLC positions in twelve arc therapy dosimetric plans of four anatomic sites: lung, pelvic, brain and head and neck. The introduced errors were from seven different amplitudes: 0.25, 0.5, 1, 2, 2.5, 3 and 5mm. The modified plans were compared to original plans by Portal Dosimetry software. The measures were analyzed by gamma analysis with 3%/3mm criteria. Results: There was no relationship between the errors amplitude and percentage of errors detected by the portal dosimetry. Comparing dosimetric plans of the same anatomical site, no significant differences in error detection were observed. Evaluating the percentage detection in different anatomical sites it was found higher proportions of detection for cases of lung and head and neck, 83.3% and 45.8%. The brain and pelvic cases had detection percentages of 12.5% and 20.8%. Complex plans as lung and H&N cases, show superior detection percentages than pelvic and brain. The 3%/3mm criteria may not be sufficiently sensitive to detect all the amplitudes of the errors that were introduced in this study particularly in simple plans. There was no significant difference in detection percentage between the leaves located in periphery and center. Conclusion: The portal dosimetry software in simple plans is less sensitive to errors in the positions of the MLC leaves, and the gamma analysis with 3%/3mm criteria does not detect most of the amplitudes used in these cases. In complex plans, the system used has an acceptable performance in terms of error detection for all the studied amplitudes. It’s important that each radiotherapy service establish appropriate protocols in accordance with the available devices and techniques practiced so as to ensure an efficient quality assurance system.
Date of Award | 15 Jun 2015 |
---|
Original language | Portuguese |
---|
Awarding Institution | - Universidade Católica Portuguesa
|
---|
Supervisor | Joana Vale Ferreira da Silva (Supervisor) |
---|
- Mestrado em Engenharia Biomédica
Controlo de qualidade em técnicas avançadas de radioterapia : RapidArc®
Silva, J. M. D. S. (Student). 15 Jun 2015
Student thesis: Master's Thesis