Room: AAPM ePoster Library
Purpose: In heavy-ion therapy, the Bragg peak precisely delivers maximum dose to the tumour at the end of the beam range, with low dose to surrounding tissue. As the Bragg peak position depends on beam energy and patient anatomy, precise range monitoring is required to ensure the prescribed dose is properly delivered. Current range monitoring techniques are often time-consuming and imprecise. In contrast, Interaction Vertex Imaging (IVI) allows sub-millimeter and real-time range monitoring under clinical conditions. Our implementation of IVI aims to further improve precision, which would allow safe delivery of higher doses to the tumour, increasing the therapeutic index.
Methods: IVI employs secondary protons created along the heavy-ion beam track in tissue. Two layers of thin silicon detectors downstream from the patient track these secondary protons. The signal from one tracker, or multiple trackers in coincidence, is filtered using our novel IVI software filter reconstruction algorithm. This algorithm reconstructs interaction vertices, the points of origin of secondary protons along the ion beam track. The resulting image of the beam is used for range monitoring.
Results: Simulations of our technique using the Geant4 Monte Carlo package with ¹²C beams of various energies allowed reconstruction of interaction vertices. Interaction vertices plotted as a function of depth produce a characteristic logistic shape close to the Bragg peak. Comparison of distal edge fit curves accurately determines differences in Bragg peak position between two fractions with sub-millimeter precision. We are currently finalizing an experimental setup to test this method using a PMMA phantom and 150 MeV/u ¹6O beam of clinical intensity at the NSCL facility, Michigan State University.
Conclusion: Simulation of our novel filtered IVI reconstruction method accurately reproduces depth difference under clinical conditions with sub-millimeter precision. First proof-of-principle experimental experiences and potential clinical applications will be presented.
Funding Support, Disclosures, and Conflict of Interest: We acknowledge the support of the CIHR, NSERC and SSHRC (under Award No. NFRFE-2018-00691, and through the USRA and CGS-M programs). This work was made possible by the facilities of the Shared Hierarchical Academic Research Computing Network (SHARCNET: www.sharcnet.ca) and Compute/Calcul Canada.