X-Ray Scatter Correction on Soft Tissue Images for Portable Cone Beam CT

Area of Science:

• Medical Imaging
• Radiological Physics
• Computational Imaging

Background:

• Soft tissue imaging using portable cone beam computed tomography (CBCT) is crucial for diagnosing conditions like tumors and intracerebral hemorrhage.
• X-ray scatter is a primary source of artifacts in CBCT, degrading image quality through phenomena like cupping artifacts, CT number inaccuracy, and reduced contrast, particularly in soft tissues.
• Existing methods struggle to adequately correct for scatter in large field-of-view CBCT systems, limiting diagnostic accuracy.

Purpose of the Study:

• To develop and evaluate an X-ray scatter correction method specifically designed to improve the quality of soft tissue images acquired with portable CBCT scanners.
• To address image degradation issues caused by X-ray scatter, thereby enhancing the potential for accurate diagnosis and detection of various pathologies.

Main Methods:

• Proposed an X-ray scatter correction method based on deconvolution using the maximum likelihood estimation maximization (MLEM) algorithm to estimate scatter signals.
• Utilized Monte Carlo simulation (MCS) software to generate scatter kernels by simulating a polymethyl methacrylate (PMMA) phantom.
• Quantitatively evaluated the method using a QRM phantom, comparing results with fan-beam CT (FBCT) data on metrics including CT number accuracy, contrast-to-noise ratio, cupping artifacts, and low-contrast detectability.
• Validated the technique on a PH3 angiography phantom to mimic human brain soft tissues.

Main Results:

• The proposed scatter correction technique demonstrated significant improvements in the quality of reconstructed soft tissue images.
• Quantitative analysis showed enhanced CT number accuracy, improved contrast-to-noise ratio, and reduced cupping artifacts compared to uncorrected images.
• The method proved effective in improving low-contrast detectability, crucial for identifying subtle abnormalities in soft tissues.

Conclusions:

• The developed X-ray scatter correction method effectively mitigates artifacts in CBCT soft tissue imaging.
• This technique shows high potential for improving the detection of soft tissues in the brain and other critical applications.
• The MLEM-based deconvolution approach offers a robust solution for enhancing diagnostic confidence in CBCT examinations.