Catheters and dose optimization using a modified CVT algorithm and multi-criteria optimization in prostate HDR...

Area of Science:

• Medical Physics
• Radiation Oncology
• Computational Biology

Background:

• Current high-dose rate (HDR) brachytherapy planning relies on planner experience for catheter positioning.
• Optimizing catheter number and location is crucial for dose coverage and reducing treatment toxicity.
• Patient-specific algorithms for catheter and dose optimization are needed.

Purpose of the Study:

• To develop and evaluate patient-specific algorithms for optimizing catheter placement and dwell times in HDR brachytherapy.
• To investigate the potential for reducing the number of catheters required while maintaining dose coverage and plan robustness.
• To assess the impact of catheter placement deviations on treatment plan quality.

Main Methods:

• A modified Centroidal Voronoi tessellation (CVT) algorithm was developed to optimize catheter positions.
• A graphics processing unit (GPU)-based multi-criteria optimization algorithm (gMCO) was used to optimize dwell times and positions.
• The algorithms were evaluated on 108 HDR brachytherapy cases for prostate cancer using real-time ultrasound, assessing plan robustness against insertion and reconstruction deviations.

Main Results:

• The combined CVT and gMCO approach increased the acceptance rate for Radiation Therapy Oncology Group (RTOG) criteria by 37% compared to standard methods.
• All patients met RTOG criteria with 11 catheters using CVT+gMCO with a 5 mm template, and 10 catheters with freehand placement.
• Plan acceptance remained high (85%) even with 3 mm deviations when using 11 catheters, with minimal decrease for 13+ catheters.

Conclusions:

• The developed CVT and gMCO algorithms enable optimization of catheter placement and dwell times for HDR brachytherapy.
• This approach can reduce the number of catheters needed for effective treatment while improving plan quality and robustness.
• Increased catheter numbers enhance plan robustness against insertion and reconstruction deviations.