Unit:
Κατεύθυνση Ιατρική Φυσική-ΑκτινοφυσικήLibrary of the School of Health Sciences
Author:
Kastoris-Katsamagkas Nikolaos
Supervisors info:
Καλέφ-Εζρά Τζων, Καθηγητής, Ιατρική Σχολή, Πανεπιστήμιο Ιωαννίνων
Θεοδώρου Κυριακή, Καθηγήτρια, Ιατρική Σχολή, Πανεπιστήμιο Θεσσαλίας
Τσέκερης Περικλής, Καθηγητής, Ιατρική Σχολή, Πανεπιστήμιο Ιωαννίνων
Original Title:
Μελέτη των συστηματικών και τυχαίων γεωμετρικών σφαλμάτων στην ακτινοθεραπεία υπό απεικονιστική καθοδήγηση (IGRT)
Translated title:
Μελέτη των συστηματικών και τυχαίων γεωμετρικών σφαλμάτων στην ακτινοθεραπεία υπό απεικονιστική καθοδήγηση (IGRT)
Summary:
The objective of this Master thesis is the study of systematic and random geometrical errors in patients subjected to image guided radiation therapy (IGRT) for 42 patients in total, with head & neck (12 patients), prostate (22 patients) and breast (8 patients) lesions. All treatments were delivered via Elekta Synergy™ linear accelerator and imaging was performed via kV XVI CBCT and MV EPID/iViewGT at Radiation Therapy Department of Ioannina Univerity Hospital. Quality control of both imaging devices was carried out and the imaging dose was measured for every imaging protocol which is in use.
The mean and the standard deviation of patients’ geometrical shifts were calculated in each dimension using the imaging devices’ software in order to register the CBCT scans to the planning CT simulator scans and the MV portal images to the DRRs and subsequently the systematic and random population errors were calculated. The PTV margins were calculated, according to the Van Herk’s formula, at 5-6mm for head & neck tumors, at 11-11,5mm for prostate tumors and at 10,5-14mm for breast tumors performing weekly imaging. The “systematic error correction” protocol was proposed considering the workload of the Department and the imaging dose. The corrections of the protocol are based on the average shift of the first three imaged fractions and then weekly imaging are conducted for re-estimating the systematic error. In addition, the use of skin tattoos instead of conventional marks on patients’ body was proposed for better alignment with the in-room lasers which resulted in the reduction of the random errors.
Furthermore, the shifts from both imaging devices were compared by Bland-Altman analysis for the anatomic region of breast. There weren’t statistically significant differences in shifts based on kV XVI CBCT and MV EPID/iViewGT, for 95% confidence intervals, at almost all cases. Therefore, it is recommended that both imaging devices are used complementary to each other so that their relative advantages are derived to the fullest.
Finally, regarding the anatomic region of prostate, geometrical errors were correlated with patients’ specific factors. A statistically significant correlation was observed between patients’ random errors on X-Lateral axis and the width of patients’ body on Z-Vertical axis, for 95% confidence intervals, at the level of isocenter.
Main subject category:
Health Sciences
Keywords:
Ακτινοθεραπεία, Απεικονιστική καθοδήγηση, Γεωμετρικά σφάλματα, Γεωμετρικά περιθώρια, CTV, PTV
