ePoster Forums
Purpose: Our proton therapy center uses a conical scintillation detector to track coincidence between radiation and imaging isocenters. The novel, four imaging-isocenter configuration of our treatment rooms necessitated an efficient means of spatially relating each unique imaging isocenter to the others and to radiation isocenter.
Methods: Custom, in-house software was developed to scrape the output files of a commercial detector system and accompanying vendor software. Raw image data scraped from the vendor software included a proton spot produced by a pencil beam incident on the detector, which contained a radio-opaque BB aligned to imaging isocenter. The relative offset of the spot signal to the BB shadow was calculated from the centroid of the respective regions of interest. Offset vectors were transformed into the proton system’s native coordinate space based on the gantry angle of delivery. The median of 2D offsets manually identified by four independent users was compared to the corresponding software-identified offsets for 84 images spanning 12 months of clinical quality assurance measurements.
Results: The average difference between the software-identified offset and the corresponding median of user-defined offsets was less than 0.1 mm in the image x- and y- directions with standard deviations of 0.1 mm and 0.2 mm for x- and y-, respectively. A paired student’s t test showed no significant difference in mean y offset (p = 0.60) and significance in mean x offset (p = 0.02) between the two identification methods.
Conclusion: Accurate and precise identification of imaging and radiation isocenters was achieved with the custom software. Quantitative and visual means of tracking multiple isocenters with the software is supported.
Quality Assurance, Protons, Image-guided Therapy
TH- External Beam- Particle/high LET therapy: Proton therapy – quality assurance