Energy layer optimization via energy matrix regularization for proton spot-scanning arc therapy.

Area of Science:

• Proton therapy physics
• Radiation oncology treatment planning
• Medical physics

Background:

• Spot-scanning arc therapy (SPArc) is an advanced proton therapy technique offering potential benefits over traditional intensity-modulated proton therapy (IMPT).
• Frequent energy layer switching (switch-up, SU) in SPArc can reduce delivery efficiency, posing a challenge for treatment planning.
• Optimizing SPArc requires balancing delivery efficiency (minimizing SU) with plan quality.

Purpose of the Study:

• To address the energy layer optimization (ELO) problem in SPArc.
• To develop a novel ELO method using energy matrix (EM) regularization.
• To enhance both plan quality and delivery efficiency for SPArc treatments.

Main Methods:

• Developed an EM regularization method to guide energy layer selection, minimizing SU while optimizing plan quality.
• Incorporated the EM into SPArc treatment planning.
• Solved the EM method using the fast iterative shrinkage-thresholding algorithm.
• Validated the EM method against the energy sequencing (ES) method.

Main Results:

• The EM method demonstrated an average of 35% reduction in SU compared to ES, improving delivery efficiency.
• EM resulted in better target dose conformity and lower doses to organs-at-risk and integral body dose.
• EM achieved at least a 10-fold improvement in computational efficiency over ES.

Conclusions:

• A new ELO method for SPArc using EM regularization has been developed.
• The EM method enhances both delivery efficiency and plan quality in SPArc.
• EM offers substantial reductions in computational time compared to the ES method.