A medical image-based graphical platform -- features, applications and relevance for brachytherapy

Area of Science:

• Medical Physics
• Radiation Oncology
• Computational Biology

Background:

• Current brachytherapy dose calculation relies on the TG-43U1 formalism.
• Emerging methods offer more accurate dose calculations by considering tissue composition, density, and geometry.
• Model-based dose calculation algorithms, particularly Monte Carlo (MC) simulations, provide a more precise approach.

Purpose of the Study:

• To present a software platform, AMIGOBrachy, for processing medical images and treatment plans.
• To prepare input data for MC simulations to improve brachytherapy dose calculation accuracy.
• To validate the AMIGOBrachy platform against established methods and clinical cases.

Main Methods:

• Developed the AMIGOBrachy graphical user interface coupled with the MCNP6 MC code for absorbed dose calculations.
• Validated AMIGOBrachy in water with an Ir-192 source and in uniform phantoms with varying materials.
• Compared MC results from patient geometries against a treatment planning system with a linear Boltzmann transport equation (LBTE) solver.

Main Results:

• TG-43U1 source parameters showed good agreement with literature values.
• MC simulations using AMIGOBrachy showed minimal dependence on tissue composition when compared to LBTE.
• Clinical cases revealed up to 25% dose differences between MC and TG-43U1, while MC and LBTE agreed within 2% for 92% of voxels.

Conclusions:

• AMIGOBrachy, utilizing MC simulations, can significantly improve the accuracy of brachytherapy dose calculations compared to TG-43U1.
• The user-friendly interface of AMIGOBrachy facilitates its integration into clinical practice.
• This platform offers a more accurate dosimetry solution for brachytherapy.