Room: Track 2
Purpose: management is a critical component of image guidance radiotherapy. We previously proposed a new scheme of using kV scattered x-ray photons for marker-less real-time image guidance in lung cancer radiotherapy and recently demonstrated its feasibility using photon counting detection technique. This study performs a Monte Carlo (MC) study to characterize its performance.
Methods: our scheme, a thin slice of x-ray beam was directed to the moving target and we measured the outgoing Compton-scattered photons using a photon counting detector with a slat collimator to establish the correspondence between pixels and scatter positions. Image corrections of geometry, beam attenuation and scattering angle were performed to convert the raw image to the actual image of Compton attenuation coefficient. We set up a MC simulation system using an in-house developed GPU-based MC package modeling the image formation process. We performed simulations for real experiments and the studies using real patient 4DCT images. We compared corrected x-ray scattering images with expected ground truth and evaluated image quality quantitatively by contrast-resolution (CR) and contrast-noise-ratio (CNR).
Results: correction, x-ray scattering image intensity and Compton scattering attenuation coefficient are linearly related, with R-squared value greater than 0.9. The CRs of raw and corrected images were 0.68 and 0.80, and the CNRs were 8.4 and 17, respectively. In the simulation for the experiments, the results were accurately reproduced with CRs of raw and corrected images being 0.75 and 0.88, and CNRs being 6.8 and 14. In the patient cases, the scattering images achieved a clear visualization of the tumor motion inside the lung. The CR and CNR were 0.63 and 3.6, respectively.
Conclusion: simulations confirmed feasibility of using scattered x-ray imaging as a method for real-time image guidance.