[P193] Improved method for in vivo dosimetry in pelvic intra-operative radiation therapy using Gafchromic films and customized digitization templates

Purpose In vivo measurements using three 1.5 × 1.5 cm 2 pieces of Gafchromic EBT3 film are part of our routine quality assurance during electron intra-operative radiation therapy (IOERT) for rectal cancer. The complexity of the pelvic irradiation geometry makes it desirable to increase the number of film samples, to ensure complete coverage of the irradiated length. The positioning of small films for digitization is time consuming and challenging, because of the need to minimize undesirable scanning artifacts. The aim of this work is to implement a practical method to digitize several small film pieces in routine practice, reducing time consumption and measuring uncertainties. Methods Three sample sizes were tested, with the same width (1.5 cm) and different length (0.75, 1.0 and 1.5 cm). Combinations of equal-sized, equally spaced samples were tested in a sacrum model for typical IOERT applicators. Film samples were digitized using a flatbed scanner Epson 10000XL. Several opaque and transparent templates for digitization were developed and tested. The uncertainties related to the films and digitization process were quantified in a previous work, and helped to determine which variations needed to be minimized (e.g. film-to-light source distance), and which could be treated more flexibly. Results The use of samples smaller than 1 × 1.5 cm 2 was discarded as impractical. Film samples of 1.5 × 1.0 cm 2 ; 1 cm apart, proved the best option for adequate coverage of the irradiated length, using strips of 3–8 samples according to applicator size. The selected template allows simultaneous digitization of a non-irradiated film piece (reference) and up to eight irradiated samples, reducing inter-scan fluctuations and total digitization time. For films irradiated with 10 to 15 Gy (dose interval of interest), this template minimizes the lateral discontinuity effect previously observed on the measured response at the films’ edges. A glass plate is used for compression, to ensure uniform film-to-light source distance. Conclusions The methodology for in vivo measurements was optimized by using smaller samples and an opaque template to read up to eight films in a single scan. These changes allow a better characterization of the irradiated area with a simpler and faster digitization method.