An integrated toolkit for structure generation, optimization, and evaluation in spatially fractionated radiation...

Area of Science:

• Medical Physics
• Radiation Oncology
• Software Engineering

Background:

• Spatially fractionated radiation therapy (SFRT) planning involves complex, unintegrated tasks: generating high-dose spheres, optimizing positions, and evaluating dose ratios.
• Current methods rely on disparate scripts and manual calculations, hindering multi-institutional SFRT research and clinical trials.

Purpose of the Study:

• To develop MAAS-SFRThelper, a unified, shared-source plugin for Varian's Eclipse treatment planning system.
• To integrate SFRT structure generation, geometric-aware optimization, and peak-valley dose ratio (PVDR) evaluation into a single workflow.

Main Methods:

• Developed as a C# WPF application using the MVVM pattern and ESAPI for Eclipse.
• Incorporates five sphere lattice generation patterns, stereotactic ablative radiation therapy (SCART) volume creation, geometric surrogate-based optimization, and multi-modal PVDR analysis.
• Validated on digital phantoms against analytic ground truth, demonstrating a 22.5% PVDR improvement in phantom testing.

Main Results:

• MAAS-SFRThelper successfully integrates SFRT planning tasks within the Eclipse environment.
• The plugin demonstrated a 22.5% increase in PVDR in phantom testing compared to unoptimized lattices.
• Outputs include new structures and summary statistics, with source code publicly available on GitHub.

Conclusions:

• MAAS-SFRThelper provides a consistent workflow for clinical lattice SFRT planning and standardized PVDR reporting for multi-institutional trials.
• Its shared-services architecture facilitates community contributions, advancing SFRT research and heterogeneous dose-delivery modalities.
• The plugin addresses the need for dosimetric consistency in SFRT, aligning with current consensus guidelines.