Last updated: May 11, 12:18 am

In this project, we implemented a quality assurance method for radiation therapy allowing to verify dose deposition in a medium externally, potentially in real time, and with no extra dose, using scattered x-ray registration. First, using the gDPM simulation package, we obtained the coordinates, momentum direction and energy of the photons scattered outside of a phantom after a MV x-ray beam passed through it. Then, using MATLAB, we collimated the scattered photons and registered them on a photon counting detector sensor. Finally, we related the recorded detector signal to the delivered radiation dose and analyzed the method feasibility. The procedures were carried out on both homogeneous and heterogeneous phantoms.

• Students: Tatiana Kashtanova, Samuel Ydenberg
• Mentor(s): Dr. Xun Jia, Dr. Lin Su (JHH)
• Technical Support: Dr. Yujie Chi (UT Arlington); Dr. Youfang Lai, Dr. Xiaoyu Hu (JHH) ​

Radiation therapy (RT) is a type of cancer treatment where a high radiation dose is delivered to the tumor while minimizing the exposure of the surrounding healthy tissue. The procedures ensuring the precise and safe delivery of radiotherapeutic dose are called Quality Assurance (QA) [1,2] and govern treatment planning , treatment delivery, and treatment delivery verification. Our project is related to the treatment delivery verification aspect of QA. During RT sessions, the greatest challenges in delivering the prescribed dose precisely come with patient mispositioning, organ movements, and anatomical morphological changes [3], all of which may result in tumor underdose and/or normal tissue overdose. The current solutions include image-guided radiation therapy (IGRT) and adaptive radiotherapy (ART) [3] where physicians use imaging to detect deviations from the initial planning and make the corresponding adjustments. The disadvantage of these techniques is an extra low radiation dose to the patient which, when accumulated over time, may result in developing secondary malignancies [4]. In this project, we implemented a quality assurance method for radiation therapy allowing to verify dose deposition in a medium externally, in real time, and with no additional dose, using scattered x-ray registration.

• Minimum: (Done on 03/28)
  1. Modeling and simulation of scattered photons registration on a photon counting detector without a true collimator present

• Expected: (Done on 04/06, 04/28, 05/02)
  1. Modeling and simulation of scattered photons registration on a photon counting detector in the presence of a collimator
  2. Modeling and programming of dose-correlated photon depth profile
  3. Project documentation

• Maximum: (Done on 04/30)
  1. Depth-dose profile retrieval in heterogeneous phantoms using scattered photons
  2. Comparison of the simulated system results with physical measurements (Excluded from the project)

The project work environment is Monte Carlo (MC) Simulation of megavoltage (MV) radiation transport into a medium. For each individual photon, the simulation takes its position, direction, and velocity as inputs, and uses a random number generator and probability distributions for different types of photon interactions with the medium to sample the distance to the next interaction [5]. Then, the process is repeated until the energy of primary and scattered photons is depleted. In the case of MV x-rays, the absorption of photons in a medium happens mainly due to the Compton process [6] where an incident photon interacts with a planetary electron having a low binding energy. The photon gives a part of its energy to the electron as kinetic energy. Then, it deflects from its original path, proceeds with reduced energy, and gets involved in further interactions. The project is based on the idea that an external registration of these Compton scattered photons might provide the information about the depth dose distribution [3, 7].

The diagram in Figure 1 depicts a MV x-ray beam applied to a water phantom. While the beam passes through the medium, many photons undergo Compton scattering. A photon counting detector with a collimator in front of it are placed next to the phantom to register the scattered photons. The collimator enables to detect only those photons that scatter at approximately right angle to the x-axis and have a certain nominal energy.

Figure 1. The diagram of RT QA method using scattered x-ray registration

The system objective is to infer on the delivered radiation dose deposition inside the phantom using information on the photons scattered outside of the phantom.

The system comprises two parts: 1) MV x-ray transport through a phantom and photons registration on a scoring (spherical) surface; 2) Scattered photons transport from the scoring sphere to a photon counting detector sensor area with and without a collimator placed in front of it. The first part must be implemented using the gDPM package [8,9] while the second part must be implemented using MATLAB. The inputs and outputs for the system parts are presented below

The system must achieve ≥ 0.9 correlation coefficient between the simulated dose and dose-correlated scattered photon profiles.

Original Dependencies

Updated Dependencies

Explanation of the changes

The dependencies were adjusted based on the refined deliverables.

Original Milestones

Updated Milestones

Explanation of the changes

The deliverables were changed per the mentors' instructions.

• Project Plan
  • Project plan presentation
  • Project plan proposal
• Project Background Reading
  • Background reading presentation
  • Background reading report
  • Simulation experiments paper [3]
  • Physical experiments paper [7]
• Project Checkpoint
  • Project checkpoint presentation
  • Poster teaser
• Project Documentation
  • System Design
  • System Requirements and Functional Specifications
  • System Testing
  • 3D Collimator Construction
• Project Code
  • GitHub Link
  • Program Structure Diagram
  • Changes in rungdpm.cu
• Project Final Presentation
  • PDF of Poster
• Project Final Report
  • Final Report
  • Project Mentors' Report

1. World Health Organization. (1988). Quality Assurance in Radiotherapy: a Guide Prepared Following a Workshop Held at Schloss Reisensburg, Federal Republic of Germany, 3-7 December1984. Geneva.

2. Glide-Hurst, C.K., & Chetty, I.J. (2014). Improving radiotherapy planning, delivery accuracy, and normal tissue sparing using cutting edge technologies. Journal of Thoracic Disease, 6(4), p.303–318.doi:10.3978/j.issn.2072-1439.2013.11.10

3. Cunha, M., et al. (2013). Dose-free monitoring of radiotherapy treatments with scattered photons: Concept and simulation study. IEEE Transactions on Nuclear Science, 60 (4), p. 3119-3126

4. Dzierma, Y., Mikulla, K., Richter, P. Bell, K., Melchior, P., Nuesken, F., & Rübe, C. (2018). Imaging dose and secondary cancer risk in image-guided radiotherapy of pediatric patients. Radiation Oncology, 13, 168.

5. The Netherlands Commission on Radiation Dosimetry. (2006). Monte Carlo treatment planning. An introduction. Report 16.

6. Hall, E.J. (2000). Radiobiology for the radiologist. 5th edition. Philadelphia, PA: Lippincott Williams & Wilkins.

7. Simões, H., et al. (2013). Dose-free monitoring of radiotherapy treatments with scattered photons: First experimental results at a 6-MV Linac. IEEE Transactions on Nuclear Science, 60(4), p. 3110-3118

8. Jia, X. & Jiang, S.B. (2011). gDPM v2.0. A GPU-based Monte Carlo simulation package for radiotherapy dose calculation. The Center for Advanced Radiotherapy Technologies (CART), UCSD.

9. Jia, X., Ziegenhein, P., & Jiang, S. B. (2014). GPU-based high-performance computing for radiation therapy. Physics in Medicine and Biology, 59(4), p. R151–R182. doi: 10.1088/0031-9155/59/4/R151