MIRD Pamphlet No. 34, Part 2: Benchmarking of MIRDct Software for CT Organ Dose Estimation

Area of Science:

• Medical Physics
• Radiological Dosimetry
• Diagnostic Imaging

Background:

• Accurate estimation of organ and effective doses in CT imaging is vital for risk assessment and protocol optimization.
• Existing CT dosimetry software varies in phantom models and input parameters, necessitating benchmarking.

Purpose of the Study:

• To systematically benchmark MIRDct, a mesh phantom-based CT dose calculation software, against NCICT and VirtualDose.
• To evaluate the agreement of organ-absorbed and effective doses calculated by MIRDct compared to reference software across diverse CT protocols.

Main Methods:

• Organ absorbed and effective doses were calculated for adult and pediatric phantoms using MIRDct, NCICT, and VirtualDose.
• MIRDct utilizes mesh-based ICRP reference phantoms, while NCICT and VirtualDose use voxel-based phantoms.
• CTDIvol values were obtained from software interfaces and scanner console outputs for 44 matched configurations.

Main Results:

• CTDIvol values showed close agreement (<7% median difference) across the three software tools.
• Organ dose coefficients agreed within ±25% for in-field organs across most adult and pediatric protocols.
• Effective dose estimates showed differences below 25% for VirtualDose (except head scans) and WB protocols, but up to 40% for NCICT chest and AP protocols.

Conclusions:

• MIRDct provides organ and effective dose estimates broadly consistent with established CT dosimetry tools.
• Agreement is typically within ±25% for in-field organs and within a few milligray for absolute doses.
• MIRDct's features support its use as a research tool for CT dose benchmarking, protocol optimization, and quality assurance.