Every year, thousands of cancer patients are treated using Intensity
Modulated Radiation Therapy (IMRT). High energy X-ray beams are
delivered to the patient from different directions, with millimeter
precision, with the objective of maximizing dosage to the tumor, while
minimizing damage to the surrounding healthy tissue. This talk is an
introduction to the basic optimization problem underlying Radiation
Treatment Planning. Specifically, we will review how the computation
of beam directions and intensities in IMRT can be formulated as a
convex optimization problem. We discuss the metrics and constraints
used in IMRT, using methods from optimization, medical physics, and
finance and risk management. We will also review some effective
parallelizable methods that have been developed for solving this
inherently large-scale, multi-objective optimization problem.
Haitham Hindi received his MS and PhD from Stanford University in Electrical Engineering, and a BSc from Imperial College in Physics. His interests are in optimization and control, and their applicationto real-world problems. He is currently a principal engineer at Walmart Labs, working on optimization algorithms for online advertising and search engine marketing. Before that, he was at Accuray, Inc, a company which makes robotic cancer radiation treatment devices. His prior work was in: energy management systems; dynamic pricing; networked and hybrid control; printing, manufacturing, and transportation networks; nonlinear control systems; particle accelerators; and disk drives.