On the Feasibility of Prompt Gamma Imaging in Heterogeneous Patient Anatomy
E Sterpin1*, G Janssens2, J Smeets2, D Prieels2, F Stichelbault2, F Roellinghoff2,3, E Clementel4, A Benilov2, S Vynckier15, (1) Universite catholique de Louvain, Center of Molecular imaging Radiotherapy and Oncology, Brussels, Belgium,(2) Ion Beam Application, Louvain-la-neuve, Belgium,(3) INSA de Lyon, Lyon, France,(4) ICTEAM institue, Louvain-la-neuve, Belgium,(5) Cliniques Univ St. Luc, Brussels, BelgiumTH-C-144-10 Thursday 10:30AM - 12:30PM Room: 144
Purpose: A prompt gamma (PG) slit camera prototype recently demonstrated 1-2 mm accuracy for the detection of range shifts of a proton pencil beam in a PMMA target. This accuracy is achieved thanks to the falloff of the prompt gamma profiles that is well correlated to the position of the Bragg peak. The present study investigates the potential of using such a camera for pencil beam scanning (PBS) in highly heterogeneous clinical situations.
Methods: PG detection profiles were simulated using a new dedicated Monte Carlo (MC) tool called 4P (PENELOPE Proton for PBS treatment Planning) previously benchmarked with the MC code MCNPX. Efficiency is enhanced in 4P by 1) neglecting neutron transport; 2) forced sampling of ICRU 63 cross sections for PG emission. Image processing and data analysis were performed on the MatLab-based REGGUI software.
In a CT scan of the thorax, incident PBS beams were defined in such a way that the protons are crossing several materials with large differences of densities (soft, bone and lung tissues). PG profiles were computed for each beam and the correlation between the profile and Bragg-peak was assessed in end-exhale CT data. The simulation was then repeated on the end-inhale phase.
Results: In general, our results show that clean PG detection profiles can be retrieved in most of tested cases. Also, we could clearly distinguish PG detection profiles computed on end-inhale and end-exhale CT data in cases where the range is modified (see Figure 1). This could lead to an on-line assessment of gating techniques. However, PG detection profiles can be greatly distorted if a beam stops right at a tumor-to-lung or lung-to-tumor interface.
Conclusion: Our current results demonstrate that PG imaging remains of interest as a strong candidate for in vivo range verification despite the presence of heterogeneities.
Funding Support, Disclosures, and Conflict of Interest: This work is partially supported by Ion Beam Application. Therefore, there is a potential conflict of interest.