Campuses:

Team 6: Interactive Treatment Planning in Cancer Radiotherapy

Monday, June 18, 2012 - 11:00am - 11:20am
Masoud Zaripisheh (University of California)
Cancer is the second cause of death in USA with estimated deaths of 570,000 in 2010. In USA, about 2/3 of cancer patients are treated with radiotherapy since it has proven a particularly effective treatment for many cancer types. Radiation is generated by a medical linear accelerator mounted on a gantry that can deliver the radiation to the patient’s body from various orientations with optimized intensity profiles of the x-ray beams (Fig-1). The main objective of radiotherapy is to deliver a lethal dose of radiation to the tumor to kill cancerous cells while sparing surrounding healthy organs and normal tissues. The treatment is complex and very patient specific. The radiation beam parameters have to be tailored to each patient's case, through a process called treatment planning, where an optimal treatment plan is designed for a particular patient based on the patient's CT image data and the physician's prescription.


Fig-1: A Medical Linear Accelerator

Fig-2: DVH Curves



(Cumulative) Dose Volume Histogram (DVH) is the most common tool employed by physicians to evaluate the quality of the plan (Fig-2). Point (D, V) on a DVH curve means for this organ, V (%) of the volume receives radiation dose more than D (Gy). Treatment planning is an multiple objective optimization problem, - delivering the desired radiation dose to the target (PTV) while minimizing dose to each healthy organ. We propose to use an interactive planning method to solve this problem. The physician will adjust the tradeoff among the target and the organs from an initial treatment plan. When the physician is satisfied with the DVH curves for some organs, s/he will lock these curves and keep looking to improve/modify the others. In this project we aim to develop a mathematically innovative and computationally efficient method to solve this important clinical problem.

Prerequisites:

Strong background in optimization, Good computing skills (MatLab or C/C++).

References:

  1. H. Edwin Romeijn and James F. Dempsey, “Intensity modulated radiation therapy treatment plan optimization,” TOP 16 (November 4, 2008): 215-243.

  2. Yong Yang and Lei Xing, “Inverse treatment planning with adaptively evolving voxel-dependent penalty scheme,” Medical Physics 31 (2004): 2839.

  3. Chuan Wu et al., “Treatment plan modification using voxel-based weighting factors/dose prescription,” Physics in Medicine and Biology 48 (August 7, 2003): 2479-2491.