Room: Room 202
Purpose: Ablative radiotherapy requires extremely precise delivery, yet tumor motion makes it difficult in practice to ensure treatment accuracy. The purpose of this study was to develop a novel technique for real-time tracking of fiducial markers using coded aperture Compton scatter imaging. This technique uses Compton scatter of the treatment beam from fiducial markers to provide live images of tumor position during treatment.
Methods: Coded aperture Compton scatter imaging was simulated in MCNPX. A 20-cm-diameter water phantom containing three gold fiducial markers was irradiated by a 4cm x 4cm 6MV field. A MURA coded aperture was placed orthogonal to the beam, and an imager was simulated along the aperture axis. A 2D high-pass filter was applied to the images in the Fourier domain, and images were reconstructed using convolution with a decoding mask. We investigated the effects of coded aperture size, distance, design, and Fourier filtration on image quality.
Results: The acquired images were dominated by scattered radiation from the water phantom. However, we observed an extremely well-defined Fourier signal from fiducial marker scatter due to the gridded nature of the coded aperture. The 2D high-pass Hamming filter was best able to filter the marker signal from the scatter background. After filtration, we were able to easily identify the fiducial marker location using a 10cm x 10cm 29-pixel aperture at a distance of 25 cm from isocenter.
Conclusion: Coded aperture imaging with Fourier filtration is well-suited to extracting the bright fiducial marker scatter signal from the broader scatter background. One can envision a clinical scenario where a coded aperture placed in front of the kV imager is used to determine live tumor position during irradiation with an MV beam. Our results demonstrate the potential feasibility of this technique to monitor live tumor motion and improve the accuracy of ablative radiotherapy.
Funding Support, Disclosures, and Conflict of Interest: This work was funded in part by the NIH K12CA086913, the University of Colorado Cancer Center/ACS IRG#57-001-53, the Boettcher Foundation, and Varian Medical Systems.