Monte Carlo investigation of sub-millimeter range verification in carbon ion radiation therapy using interaction vertex imaging

Devin Hymers,Dennis Mücher
DOI: https://doi.org/10.1088/2057-1976/aafd44
2019-06-26
Abstract:In hadrontherapy, particularly carbon ion radiation therapy, a characteristic dose distribution precisely delivers maximum dose at the beam endpoint, or Bragg peak, with relatively low dose to the surrounding tissue. As the position of the Bragg peak is highly dependent on patient anatomy and physiology, precise range verification techniques are needed to ensure that the prescribed dose is properly targeted to tumours while sparing healthy tissue. We simulated treatments of a homogeneous phantom using Geant4, and applied a novel Interaction Vertex Imaging (IVI) reconstruction, combining single-particle reconstruction with a coincidence technique, and using a software filter to reduce uncertainty introduced by straggling and multiple scattering in the target. Interaction vertices generated by the most precise Triangulation IVI method were localized to an average of 3.5 mm from the true position of the reaction, simulating a realistic charged particle detection system. No event-by-event information from a beam tracking detector was used in reconstruction. Filtered longitudinal vertex distributions were fit to logistic functions, characterizing the distal edge closest to the Bragg peak. Comparing the position of this distal edge between simulations allowed us to accurately determine if two treatments correctly targeted the same depth. After performing a linear calibration, the depth difference between two treatments could be determined with sub-millimeter precision under clinical conditions (106–107 incident 12C ions), allowing range verification to be performed for each depth setting in a pencil beam scanned treatment plan.
What problem does this paper attempt to address?