Related Experiment Video
Updated: Jul 18, 2026

Proton Therapy Delivery and Its Clinical Application in Select Solid Tumor Malignancies
Published on: February 6, 2019
Optimization of HDR brachytherapy dose distributions using linear programming with penalty costs
Ron Alterovitz1, Etienne Lessard, Jean Pouliot
1Department of Industrial Engineering and Operations Research, University of California, Berkeley, 4141 Etcheverry Hall, Berkeley, California 94720-1777, USA. ron@ieor.bankeley.edu
This study reformulates prostate cancer radiotherapy planning, using linear programming (LP) for high-dose-rate (HDR) brachytherapy. The new LP method finds optimal radioactive source dwell times faster and more accurately than traditional simulated annealing (SA).
Area of Science:
- Medical Physics
- Radiation Oncology
- Computational Optimization
Background:
- High-dose-rate (HDR) brachytherapy is a key treatment for prostate cancer.
- Accurate dose distribution requires precise optimization of radioactive source dwell times within implanted catheters.
- Current methods like Inverse Planning by Simulated Annealing (IPSA) use probabilistic approaches with potential for suboptimal solutions.
Purpose of the Study:
- To reformulate the HDR brachytherapy dwell time optimization problem as a linear programming (LP) problem.
- To compare the performance and solution quality of the new LP method against the established simulated annealing (SA) approach within IPSA.
Main Methods:
- The established Inverse Planning by Simulated Annealing (IPSA) dwell time optimization problem was exactly formulated as a linear programming (LP) problem.
- The LP formulation was applied to a cohort of 20 prostate cancer patient cases.
- LP-derived dwell times were quantitatively compared against dwell times generated by the clinical SA method.
Main Results:
- The LP method achieved globally optimal dwell time solutions, minimizing the IPSA objective function.
- LP optimization was rapid, completing in under 15 seconds per case on a standard PC.
- LP solutions yielded significantly improved objective function values compared to SA (P = 1.54 x 10(-7)).
Conclusions:
- Linear programming provides an exact and deterministic method for optimizing dwell times in HDR prostate brachytherapy.
- The LP approach offers superior objective function values compared to SA while maintaining clinical equivalence in dosimetric indices.
- This LP formulation enhances treatment planning by ensuring mathematically optimal solutions with strong performance guarantees.
More Related Videos
Related Concept Videos
Lagrange Multipliers: One Constraint
Lagrange Multipliers: Problem Solving
Lagrange Multipliers: Two Constraints
Optimization Problems
Determination of Multiple Dosing Parameters: Loading and Maintenance Doses

