POLY2TET: a computer program for conversion of computational human phantoms from polygonal mesh to tetrahedral mesh

Area of Science:

• Medical Physics
• Computational Biology
• Radiation Dosimetry

Background:

• Tetrahedral mesh (TM) offers advantages over polygonal mesh (PM) for computational human phantoms, including better compatibility with Monte Carlo radiation transport codes and faster computation.
• Existing PM phantoms present limitations in modeling heterogeneous densities and integrating with certain simulation tools.

Purpose of the Study:

• To develop a computer program, POLY2TET, for converting polygonal mesh (PM) human phantoms to tetrahedral mesh (TM) format.
• To enable the use of TM phantoms in general-purpose Monte Carlo radiation transport codes such as Geant4, PHITS, and MCNP6.
• To evaluate the performance of converted TM phantoms in terms of computation speed, memory requirements, and dose calculation accuracy.

Main Methods:

• Developed the POLY2TET program to convert PM phantoms to TM format, generating necessary source codes or input files for target Monte Carlo codes.
• Tested the POLY2TET program using four high-fidelity PM human phantoms.
• Compared the computational speed, memory usage, and initialisation time of TM phantoms against original PM phantoms within the Geant4 simulation environment.

Main Results:

• The POLY2TET program successfully converted PM phantoms to TM format.
• Organ doses calculated using TM phantoms in Geant4, PHITS, and MCNP6 were comparable to those obtained with original PM phantoms.
• TM phantoms demonstrated significantly faster particle transport, up to approximately 2600 times faster for electron beams in PHITS.
• TM phantoms generally required more memory than PM phantoms, but remained within typical personal computer limits (<12 GB) for Geant4 and PHITS, though MCNP6 required substantially more (60-70 GB).

Conclusions:

• The developed POLY2TET program effectively converts PM human phantoms to TM format, facilitating their use in major Monte Carlo radiation transport codes.
• TM phantoms offer substantial computational speed advantages for radiation transport simulations, particularly with electron beams.
• While memory requirements for TM phantoms can be higher, they are often manageable for common simulation tasks, with MCNP6 being a notable exception.