SU-F-BRCD-04: Efficient Scatter Distribution Estimation and Correction in CBCT Using Concurrent Monte Carlo Fitting.

Area of Science:

• Medical Imaging
• Computational Imaging
• Radiology

Background:

• X-ray scatter is a significant source of artifacts and reduced image quality in cone-beam CT (CBCT).
• Accurate scatter correction is crucial for reliable quantitative analysis and diagnostic accuracy in CBCT imaging.
• Existing scatter correction methods may face limitations in computational efficiency or accuracy.

Purpose of the Study:

• To develop and evaluate a novel, computationally efficient scatter correction method for CBCT.
• To demonstrate the performance of an algorithm that combines multiple Monte Carlo (MC) CBCT scatter simulations with a fitting function.
• To assess the impact of the proposed scatter correction on image quality metrics.

Main Methods:

• The method employs concurrently run MC CBCT scatter projection simulations at a subset of projection angles.
• A fitting function, specifically a frequency-limited sum of sines and cosines, is used to estimate the scatter detector response.
• The estimated scatter is subtracted from the original projections to generate scatter-corrected reconstructions.

Main Results:

• The scatter estimation and correction were achieved in under one minute for a 360-degree projection set.
• The algorithm demonstrated a 46% increase in contrast-to-noise ratio.
• Significant reductions in artifacts were observed, including an 87% decrease in shading artifact and a 79% decrease in skin line artifact in a simulated pelvis phantom.

Conclusions:

• The developed algorithm provides an efficient and effective method for scatter estimation and removal in CBCT.
• The proposed scatter correction technique significantly enhances image quality in CBCT reconstructions.
• This method holds promise for improving diagnostic accuracy and quantitative measurements in CBCT applications.