Student(s): Aderet Barak, Griffin Moore, Kyle Smith
Project: Radiation Therapy Optimization
Advisor(s): Tamas Terlaky, Mohamad Mohammadisiahroudi
Abstract
Radiation therapy treatment planning requires solving complex optimization problems to determine how radiation dose should be distributed to eradicate a tumor while minimizing harm to healthy tissues. Intensity-modulated proton therapy (IMPT), in contrast to conventional photon radiation therapy, delivers a high radiation dose at specific locations along the beamlet path. While IMPT enables more precise dose delivery, it also leads to complex treatment planning optimization problems that must simultaneously satisfy multiple prescribed dose-volume constraints. These requirements often result in large-scale mixed-integer optimization problems. The computational burden associated with solving these problems grows substantially, limiting scalability and restricting the ability to explore globally optimal solutions within clinically acceptable time frames. This project investigates the quantum hamiltonian descent (QHD) algorithm as a pathway toward more computationally efficient IMPT planning. By reformulating dose constraints and beamlet weights into mixed-integer models compatible with quantum optimization frameworks, we evaluate whether QHD implementations can improve runtime performance relative to conventional classical methods, while providing comparable efficacy. The QHD pipeline supports both execution on real quantum hardware and GPU-accelerated classical simulation. Results from the GPU-accelerated implementation demonstrate a speedup compared with the state-of-the-art classical solver Gurobi. In addition, compared with current clinical systems, the procedure is largely automated and requires minimal input and intervention from planners, which helps accelerate the clinical workflow. A broader impact of this approach is that it facilitates robust optimization to generate treatment plans that are resilient to patient motion, which is one of the major challenges in IMPT.
About Aderet Barak
Major: Industrial and Systems Engineering
Aderet Barak is a junior at Lehigh University pursuing a B.S. in Industrial and Systems Engineering with a minor in Data Science. Her research focuses on proton therapy treatment planning, where she develops optimization models to improve radiation dose delivery while minimizing exposure to healthy tissue. Under faculty mentorship, she applies robust and chance-constrained optimization techniques to address uncertainty in clinical parameters and enhance treatment reliability. Aderet is particularly interested in leveraging mathematical modeling and data-driven methods to solve complex healthcare and operational challenges. In addition to her research, she has experience in continuous improvement initiatives and is completing a summer role in the pharmaceutical industry, where she applies analytical and process optimization skills in a real-world setting.
About Griffin Moore
Major: Computer Science and Engineering
Griffin Moore is a second-year student at Lehigh University pursuing a bachelor’s degree in Computer Science and Engineering. This is his second semester working on the Radiation Therapy Optimization project in the Industrial and Systems Engineering department at Lehigh. There, he formulates optimization problems for radiation therapy treatment plans in order to compare the results of solving them with classical solvers to those of the Quantum Hamiltonian Descent solver. Prior to that, Griffin took part in an independent study in quantum computing, working to outline a lesson plan to guide other students in a similar independent study or potentially a course on the same material. Griffin intends to pursue a career of research in quantum computing and quantum information, specifically in optimization and modeling physical systems.
About Kyle Smith
Major: Computer Science and Business
Kyle Smith is a senior studying Computer Science and Business at Lehigh University whose work sits at the intersection of computation and healthcare. His current research focuses on optimizing radiation therapy treatment planning using advanced mixed-integer modeling and quantum-inspired optimization techniques. He explores how emerging computational approaches can improve dose distribution decisions in proton therapy, with the goal of delivering more precise cancer treatment. Kyle’s interest in healthcare innovation was shaped by his experience conducting fieldwork in the Philippines through the Global Social Impact Fellowship, where he helped develop a digital system to improve tuberculosis treatment adherence. That experience reinforced his belief that better systems and smarter algorithms can improve patient outcomes. He is particularly interested in the potential of quantum computing to solve large-scale medical optimization problems and expand access to efficient, data-driven healthcare solutions.