3D Proton Radiography of Simple Inhomogeneities: Implications for Tumor Depth Tracking
M Raytchev*, J Seco, Mass General Hospital; Harvard Medical School, Boston, MASU-D-134-1 Sunday 2:05PM - 3:00PM Room: 134
Purpose: To develop a method to determine the depth and thickness of simple geometry inhomogeneities from single proton radiographic images. The final goal was to provide a methodology to allow 3D tumor tracking in real-time.
Methods: Inhomogeneous tissues spanning the densities from bone to lung were placed within a water phantom. Their location and thickness were parameterized to build an exhaustive 2D array of phantoms. Monte Carlo simulations with customized scoring of both energy fluence and track count ratios were performed to build 2D isofurfaces over the 2D inhomogeneity parameter space. Both beams passing through the inhomogeneity and beams not pass through the inhomogeneity were scored by each scorer to form these ratios. The coordinates of the crossing point between the isolines determined over both the track count ratio and the energy fluence ratio isosurfaces at levels corresponding to these ratios for an unknown structure could then be used to recognize the depth and thickness of the unknown inhomogeneity.
Results: To validate the method throughout the parameter space the recognition of each phantom was tested. Inhomogeneity depth and location recognition was unique for almost all test phantoms when the inhomogeneity material and phantom thickness were known versus generally less than 10 false positives otherwise. The errors in recognitions were confined to the edges of the isosurfaces. While the method is currently not ready for clinical use, a resolution improvement to 1.5-2 mm was estimated to require acceptable dose to the patient. Preliminary analysis indicates that the influence of the edge effects around tissue interfaces and the complex anatomy could potentially be accounted for within the framework of this reference method.
Conclusion: A method implementing 3D proton radiography was developed and evaluated. Further development will be required to achieve clinically relevant 3D tumor tracking.
Funding Support, Disclosures, and Conflict of Interest: The project was supported by the Federal Share of program income earned by Massachusetts General Hospital on C06 CA059267, Proton Therapy Research and Treatment Center. The authors declare no conflict of interest.