Room: Room 202
Purpose: The long term goal of the project is to image the dose and range from a proton beam during the treatment of brain cancer in pediatric patients. In this study, the effects of the skull on the 3D RACT images is evaluated, where changes in sensitivity, positional accuracy, and image distortion are simulated.
Methods: The thermoacoustic signals from an array of 2mm diameter absorbers within a spherical skull was simulated. The physical parameters of the skull were obtain in the literature for both adult skulls and children (3.5-mm thick), including the inner and outer compact bone and diploe; and the attenuation, speed of sound, and acoustic impedance as a function of frequency. A ray tracing algorithm was used to measure the effects of refraction, transmission losses, and attenuation of the thermoacoustically-generated pressure through the skull. Implementing a prototype RACT scanner design (72 transducer array on a rotating cylinder) and a 3D filtered-backprojection algorithm with speed-of-sound corrections, the array of absorbers was reconstructed. The maximum intensity of each spherical absorber was compared to the intensity without the skull, as was the sphereâ€™s location and diameter.
Results: For a pediatric skull, the image intensity decreased by 1/3 at isocenter and 1/10 near the edge of the skull. The positional accuracy of the spheres remained less than 0.2mm, while the diameter perpendicular to the radial vector increased to 3.0-3.5mm when placed near the skullâ€™s edge.
Conclusion: Overall, the loss of projections due to long attenuation pathlengths resulted in a significant amount of distortion and a 60-90% loss in sensitivity, without loss in positional accuracy. In pediatric patients, this is anticipated to result in a single pulse sensitivity in the 4.8-16cGy range, based on past RACT simulations. Iterative techniques will be required to reduce distortions near the periphery.