Spherical cluster analysis for beam angle optimization in intensity-modulated radiation therapy treatment planning

Area of Science:

• Medical Physics
• Radiation Oncology
• Computational Biology

Background:

• Intensity-modulated radiation therapy (IMRT) planning involves selecting optimal beam angles.
• Current methods may not fully leverage non-coplanar beams for complex cases.
• Patient-specific factors influence the ideal beam configuration.

Purpose of the Study:

• To introduce and validate a novel heuristic for establishing IMRT beam configurations.
• To assess the clinical impact of optimized non-coplanar beam ensembles compared to standard plans.
• To evaluate the method's effectiveness across different anatomical sites (intra-cranial, pancreas, prostate).

Main Methods:

• A patient-specific matrix assessed radiological quality for candidate beam directions per target voxel.
• A spherical K-means algorithm identified clusters of locally ideal beam angles.
• Optimized beam ensembles were used in conventional inverse planning for fluence profile optimization.
• Computation time for beam ensemble pre-optimization averaged six minutes.

Main Results:

• Optimized non-coplanar nine-beam ensembles yielded clinically comparable dose distributions for prostate cases.
• For pancreas and intra-cranial cases, optimized ensembles improved organ-at-risk sparing while maintaining target coverage.
• Symmetric patient geometries showed minimal improvement with optimized beams over equi-spaced coplanar beams.
• The method demonstrated significant benefit for asymmetrically located targets near organs at risk.

Conclusions:

• The proposed beam angle optimization strategy, using spherical cluster analysis, is effective for IMRT.
• Optimized non-coplanar beams offer advantages in sparing organs at risk for specific patient anatomies.
• This heuristic provides an intuitive approach to beam ensemble selection, particularly for challenging treatment sites.