Optimal Multileaf Collimator Leaf Sequencing in IMRT Treatment Planning

Authors:
Z. Caner Taşkın;J. Cole Smith;H. Edwin Romeijn;James F. Dempsey
Affiliations:
Department of Industrial and Systems Engineering, University of Florida, Gainesville, Florida 32611;Department of Industrial and Systems Engineering, University of Florida, Gainesville, Florida 32611;Department of Industrial and Operations Engineering, University of Michigan, Ann Arbor, Michigan 48109;Department of Radiation Oncology, University of Florida, Gainesville, Florida 32610, and ViewRay Inc., Village of Oakwood, Ohio 44146
Venue:
Operations Research
Year:
2010

Citing 10
Cited 4

Optimizing the Delivery of Radiation Therapy to Cancer Patients

SIAM Review
Inverse radiation therapy planning: a multiple objective optimization approach

Discrete Applied Mathematics - Special issue: Third ALIO-EURO meeting on applied combinatorial optimization
A coupled column generation, mixed integer approach to optimal planning of intensity modulated radiation therapy for cancer

Mathematical Programming: Series A and B
Minimizing beam-on time in cancer radiation treatment using multileaf collimators

Networks
A network flow algorithm to minimize beam-on time for unconstrained multileaf collimator problems in cancer radiation therapy

Networks
A Column Generation Approach to Radiation Therapy Treatment Planning Using Aperture Modulation

SIAM Journal on Optimization
A New Linear Programming Approach to Radiation Therapy Treatment Planning Problems

Operations Research
Decomposition of integer matrices and multileaf collimator sequencing

Discrete Applied Mathematics
A new algorithm for optimal multileaf collimator field segmentation

Discrete Applied Mathematics
A duality based algorithm for multileaf collimator field segmentation with interleaf collision constraint

Discrete Applied Mathematics

A note on improving the performance of approximation algorithms for radiation therapy

Information Processing Letters
A shortest path-based approach to the multileaf collimator sequencing problem

Discrete Applied Mathematics
A function approximation approach to the segmentation step in IMRT planning

OR Spectrum
Hybrid methods for the multileaf collimator sequencing problem

CPAIOR'10 Proceedings of the 7th international conference on Integration of AI and OR Techniques in Constraint Programming for Combinatorial Optimization Problems

Quantified Score

Hi-index	0.00

Visualization

Abstract

We consider a problem dealing with the efficient delivery of intensity modulated radiation therapy (IMRT) to individual patients. IMRT treatment planning is usually performed in three phases. The first phase determines a set of beam angles through which radiation is delivered, followed by a second phase that determines an optimal radiation intensity profile (or fluence map). This intensity profile is selected to ensure that certain targets receive a required amount of dose while functional organs are spared. To deliver these intensity profiles to the patient, a third phase must decompose them into a collection of apertures and corresponding intensities. In this paper, we investigate this last problem. Formally, an intensity profile is represented as a nonnegative integer matrix; an aperture is represented as a binary matrix whose ones appear consecutively in each row. A feasible decomposition is one in which the original desired intensity profile is equal to the sum of a number of feasible binary matrices multiplied by corresponding intensity values. To most efficiently treat a patient, we wish to minimize a measure of total treatment time, which is given as a weighted sum of the number of apertures and the sum of the aperture intensities used in the decomposition. We develop the first exact algorithm capable of solving real-world problem instances to optimality within practicable computational limits, using a combination of integer programming decomposition and combinatorial search techniques. We demonstrate the efficacy of our approach on a set of 25 test instances derived from actual clinical data and on 100 randomly generated instances.