Purpose: In image guided brachytherapy for cervical cancer the dose accumulation is based on a worst case scenario assumption that the anatomical position of the hot spots in the organs at risk remains constant for each fraction. The purpose of this study was to determine the reproducibility of three different methods of deformable image registration (DIR) and their effect on cumulative organ at risk dose volume histogram (DVH) parameters summed over three brachytherapy fractions and to assess the effect of DIR on the anatomical position of the DVH parameters for bladder and rectum. Materials and Methods: DIR was retrospectively completed for 39 cervical cancer patients based on CT planning images from three brachytherapy fractions. DIR was completed using three commercially available methods within MIM Software (MIM Software Inc., USA) 1) unconstrained deformation of the entire image, 2) using bladder contour based DIR, 3) masking the bladder with 1000 hounsfield units (HU) before whole image DIR. Locally locked adjustments were used to refine the DIR. Registrations were repeated using each brachytherapy image as the primary image to assess the reproducibility. Accumulated DVH parameters (D2cc and D0.1cc) were calculated for both bladder and rectum and compared with the dose summation without using DIR. Based on the original fraction 3 image, structure set and dose map, structures representing the D2cc and D0.1cc were created from the overlap between the DVH parameter isodose and the bladder or rectum structure. This was repeated for the same image and structure set using the deformed dose map, accumulating the doses from the three fractions. The DICE index was used to calculate the overlap of the D2cc and D0.1cc structures from the original fraction 3 and deformed accumulated dose maps. Results: Least variation in bladder D2cc and D0.1cc was observed when using DIR method 3. With this DIR method and accumulating dose on the primary image from fraction 3, the D2cc and D0.1cc decreased by 2.86% and 4.18% respectively compared to the summed dose with no DIR. The average DICE index for the bladder was 0.778 and 0.613 for the D2cc and D0.1cc respectively. For the rectum, using DIR method 3, the average D2cc decreased by 5.08% and the average D0.1cc decreased 2.77%. The average DICE was 0.825 and 0.622 for D2cc and D0.1cc respectively. Conclusions: Masking the bladder contour with a high HU prior to DIR gave the most consistent results. Significant dose decreases were observed for summed DVH parameters over the three fractions. The final fraction summed with deformed doses from previous fractions demonstrated hot spots in different locations to the original position for the final fraction. Further work needs to be undertaken to determine the clinical impact of DIR use in DVH parameter addition for OAR in HDR brachytherapy for cervical cancer.
PO06Ultrasound Image Based Brachytherapy Planning for Carcinoma Cervix-Feasibility Study
Removal of the flattening filter alters the energy spectrum of the photon beam such that current beam quality specifiers may not correctly account for this change when predicting the Spencer-Attix restricted water-to-air mass collision stopping-power ratio, (L/ρ)(water)(air). Johnsson et al (2000 Phys. Med. Biol. 45 2733-45) proposed a beam quality specifier, known as the dual parameter beam quality specifier, which was calculated via Monte Carlo (MC) simulations using transmission data of primary kerma through two differing thicknesses of water material. Ceberg et al (2010 Med. Phys. 37 1164-8) extended this MC study to include relevant flattening filter free (FFF) beam data. Experimental investigations of this dual parameter beam quality specifier have not previously been published, therefore the purpose of this work was to validate that the dual parameter beam quality specifier could be measured experimentally for clinical beams (both with a flattening filter (WFF) and without (FFF)). Transmission measurements of primary kerma were performed by employing the setup outlined in Johnsson et al (1999 Phys. Med. Biol. 44 2445-50). Varying absorber thicknesses, in 5 cm increments from 5 to 40 cm, were placed at isocentre with the chamber positioned at an extended source to chamber distance of 300 cm. Experimental setup for TPR20,10 and %dd(10)x followed the methodology outlined in IAEA TRS398 (2004) and TG-51 (1999) with AAPM Addendum to TG-51 (2014) respectively. The maximum difference of (L/ρ)(water)(air) determined using the different beam quality specifiers was found to be 0.35%. Analysis of the absorber thickness combination found that small thicknesses (<10 cm) for the first absorber and absorbers similar in thickness (<10 cm) should be avoided. Stopping-power ratios of the beams investigated were determined using three different beam quality specifiers. The results demonstrated successful experimental determination of the dual parameter beam quality specifier, indicating its potential as an alternate beam quality specifier for FFF beams.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.