A Probability-Based Monte Carlo Sampling Algorithm for Visualizing X-Ray Compton Scattering in Virtual Reality – UROP Symposium

A Probability-Based Monte Carlo Sampling Algorithm for Visualizing X-Ray Compton Scattering in Virtual Reality

Andy Tang

Research Mentor: Kimberlee Kearfott
Mentor Department: NERS, Engineering
Author(s): Andy Tang, Jack Drougel, Ava Geisler, Kimberlee Kearfott
Session: Session 6 (3:00 PM – 3:50 PM)
Presentation Type: Poster 64

Abstract

Radiation interactions such as Compton scattering are important in medical imaging and radiation health. However, they can be difficult to visualize and understand. Many educational materials rely on static 2D diagrams or simplified animations. These approaches often do not capture the three-dimensional and probabilistic nature of scattering. As a result, learners may struggle to connect the underlying physics to real processes in X-ray imaging, where scattered photons can reduce image quality. This project addresses that gap by adding an interactive Compton scattering simulation to the Tiny Tours virtual reality environment. Tiny Tours is designed to support first-person, hands-on exploration of radiation concepts. A probability-based Monte Carlo sampling algorithm was implemented using rejection sampling and the Klein-Nishina equation to generate scattering angles from incident photon energy. The simulation models a single photon interacting with a single electron and updates the photon’s scattered direction and energy in real time. Simple 3D visual elements are used to show how scattering behavior changes across energy levels without unnecessary visual clutter. Supporting X-ray-related assets were also developed to place the interaction in a broader medical imaging context. The expected result is that repeated sampling will produce angular distributions consistent with theoretical Klein-Nishina predictions. This will be evaluated by comparing sampled angles with the expected distribution and by checking performance during real-time interaction. Overall, this work shows how Monte Carlo-style probabilistic simulation and interactive graphics can make abstract radiation physics concepts more intuitive and accessible.

lsa logoum logo