An efficient quasi-Monte Carlo method with forced fixed detection for photon scatter simulation in CT

Area of Science:

• Medical Imaging
• Computational Physics
• Radiological Sciences

Background:

• Scattered photons in CT imaging cause cupping and streak artifacts, degrading image quality.
• Accurate estimation of scatter intensities is crucial for artifact reduction in CT scans.

Purpose of the Study:

• To develop a fast and accurate algorithm for estimating scatter intensities in CT imaging.
• To improve the quality of CT images by mitigating artifacts caused by scattered photons.

Main Methods:

• Transformed photon path probability into a high-dimensional integral.
• Developed gQMCFFD: a GPU-based quasi-Monte Carlo (QMC) algorithm with forced fixed detection.
• Utilized low discrepancy sequences for deterministic Monte Carlo simulations.

Main Results:

• gQMCFFD achieved efficiency improvement factors of 4-46 times over GPU-based Monte Carlo methods.
• Combined gQMCFFD with sparse matrix methods for rapid simulation (2 seconds per projection angle).
• Achieved a relative difference of 3.53% in scatter intensity estimation.

Conclusions:

• gQMCFFD offers a significant advancement in fast and accurate scatter intensity estimation for CT imaging.
• The algorithm effectively reduces artifacts, leading to enhanced CT image quality.
• This method shows promise for real-time or near-real-time CT image correction.