Encrypted login | home

Program Information

Enhancement of Lung Tumor Visibility by Dual-Energy X-Ray Imaging in An Anthropomorphic Chest Phantom Study

no image available
M Menten

MJ Menten*, MF Fast , S Nill , U Oelfke , The Institute of Cancer Research and The Royal Marsden NHS Foundation Trust, London, UK

Presentations

WE-E-18A-2 Wednesday 1:45PM - 3:45PM Room: 18A

Purpose: Intrafractional lung tumor motion during radiotherapy can be compensated for by tracking the tumor position using x-ray imaging and adapting the treatment in real-time. However, locating the tumor with an automated template-matching algorithm is often challenging if the tumor is obscured by ribs. This study investigates the feasibility of creating dual-energy (DE) images of the chest with increased tumor visibility on an Elekta XVI system.
Methods: An anthropomorphic chest phantom was imaged at two different energies. Low-energy images were obtained at 80 kVp (0.8 mAs); high-energy images at 129 kVp (0.6 mAs, additional 1.26 mm tin filter). A Geant4 Monte-Carlo framework was developed allowing simulation of the x-ray tube, flat-panel detector and phantom in order to optimize the beam energies, filtration and the weighting factor used to subtract the individual images into a synthetic DE image. The weighting factor was selected to minimize the visibility of bones while maintaining a sufficient tumor visibility. We scored the bone visibility as the contrast of tumor (with bone) to tumor (without bone), and similarly of lung tissue (with bone) to lung tissue (without bone). Tumor visibility was quantified as the contrast between tumor and lung tissue (both without bone).
Results: In the experimentally obtained DE image the bone visibility was reduced by 79.2% in tumor and by 96.8% in lung tissue while the overall tumor visibility only decreased by 69.5%. The Monte-Carlo simulation yielded similar results reducing the scores by 90.0%, 85.3% and only 71.9%, respectively.
Conclusion: This work demonstrates the feasibility of DE imaging to enhance lung tumor detectability. In the future, we hope to further refine the Monte-Carlo simulation to more accurately predict the weighting factors which would aid real-time implementation. Furthermore, we plan to use the Monte-Carlo framework to simulate DE images of actual lung tumors.



Funding Support, Disclosures, and Conflict of Interest: The authors would like to thank Paul-Scherrer-Institut and Centre Suisse d'Electronique et Microtechnique for lending us the anthropomorphic phantom. We acknowledge support from Elekta AB under a research agreement. Research at The Institute of Cancer Research is also supported by Cancer Research UK under Programme C46/A10588.


Contact Email: