A novel reduced-order prioritized optimization method for radiation therapy treatment planning

Area of Science:

• Medical Physics
• Computational Biology
• Radiotherapy

Background:

• Radiation therapy treatment planning involves complex optimization problems.
• Balancing target coverage and organ-at-risk sparing is crucial.
• Existing methods may face computational challenges in high-dimensional intensity spaces.

Purpose of the Study:

• To introduce a novel reduced-order prioritized algorithm for radiation therapy treatment planning.
• To improve the efficiency and effectiveness of intensity-modulated radiation therapy (IMRT) planning.
• To reduce computational time while maintaining or improving plan quality.

Main Methods:

• A three-stage approach: intensity space sampling, principal component analysis for dimensionality reduction, and sequential prioritized optimization.
• Utilizing Latin hypercube sampling for weight definition in unconstrained problems.
• Implementing a slip factor to manage planning target volume (PTV) coverage constraints.

Main Results:

• Demonstrated applicability in prostate and lung IMRT cases.
• Achieved significant mean dose reductions in rectum (21.3%) and bladder (22.4%).
• Reported speed-up factors up to 49.9 due to dimensionality reduction with minimal PTV dose degradation (~4%).

Conclusions:

• The reduced-order prioritized algorithm offers a computationally efficient solution for radiation therapy planning.
• The method effectively reduces dose to organs at risk while preserving target coverage.
• This approach shows promise for enhancing IMRT treatment planning workflows.