A Monte Carlo-based model for simulation of digital chest tomosynthesis

Area of Science:

• Medical Imaging
• Computational Physics
• Radiological Sciences

Background:

• Digital chest tomosynthesis (CT) is an advanced imaging technique.
• Accurate simulation models are crucial for optimizing CT systems and image quality.
• Realistic simulation requires incorporating factors like scatter and noise.

Purpose of the Study:

• To develop and validate a Monte Carlo-based computer simulation model for generating synthetic digital chest tomosynthesis projections.
• To adapt the model to simulate the GE Healthcare VolumeRAD system.
• To enable the optimization of existing and future chest tomosynthesis systems.

Main Methods:

• An anthropomorphic chest phantom was scanned, segmented, and incorporated into a computer model.
• Sixty tomosynthesis projections were simulated using the Monte Carlo method with specific input parameters.
• Image reconstruction was performed using noise-free, scatter-included, and noise-added projections for comparison.

Main Results:

• The simulation model successfully generated realistic, high-resolution X-ray images.
• Reconstructed images from simulated projections were compared to those from a real phantom.
• The model demonstrated the capability to simulate various noise and scatter conditions.

Conclusions:

• The developed Monte Carlo model provides a viable tool for simulating chest tomosynthesis.
• This simulation approach aids in the optimization of tomosynthesis systems and protocols.
• Further development aims to enhance the model for broader applications in medical imaging research.