A DVH-guided IMRT optimization algorithm for automatic treatment planning and adaptive radiotherapy replanning.

Area of Science:

• Medical Physics
• Radiation Oncology
• Computational Biology

Background:

• Intensity modulated radiation therapy (IMRT) optimization is complex, requiring significant clinician expertise.
• Automatic treatment planning and adaptive radiotherapy (ART) replanning aim to improve efficiency and consistency.
• Incorporating prior treatment knowledge can guide optimization and improve plan quality.

Purpose of the Study:

• To develop a novel algorithm for IMRT optimization that integrates prior treatment knowledge.
• To facilitate automatic treatment planning and ART replanning.
• To enable automatic adjustment of voxel weights based on reference plan DVH curves.

Main Methods:

• Developed a voxel-based optimization algorithm using reference plan DVH curves to guide IMRT.
• Algorithm navigates the Pareto surface, iteratively adjusting voxel weights.
• Tested on three patient cases, with GPU implementation for efficiency.

Main Results:

• The algorithm successfully generated Pareto optimal plans with DVH trade-offs similar to reference plans.
• Demonstrated automatic adjustment of voxel-weighting factors.
• Achieved high efficiency through GPU implementation.

Conclusions:

• A novel prior-knowledge-based optimization algorithm for IMRT has been developed.
• The algorithm efficiently generates clinically optimal plans with improved quality.
• Significant improvements in planning efficiency for ART replanning and automatic treatment planning were observed.