Image artifacts and noise reduction algorithm for cone-beam computed tomography with low-signal projections

Area of Science:

• Medical Imaging
• Image Processing
• Computational Science

Background:

• Cone-beam computed tomography (CBCT) often suffers from low-signal projections, leading to image artifacts and noise.
• Existing reconstruction algorithms struggle to produce high-quality images from noisy or incomplete data.

Purpose of the Study:

• To develop and evaluate a novel image reconstruction algorithm for CBCT.
• To improve image quality by effectively reducing artifacts and noise in low-signal projection data.

Main Methods:

• A multiple sampling method was used in the projection domain to suppress environmental noise.
• A fuzzy entropy-based method combined with Block Matching 3D (BM3D) filtering was applied in the image domain.
• The algorithm's performance was assessed through simulation studies using a polychromatic spectrum.

Main Results:

• The new algorithm significantly improved signal-to-noise ratios (SNRs) and contrast-to-noise ratios (CNRs).
• Reconstructed images showed an average 40% increase in SNRs and a 20% increase in CNRs compared to traditional methods.
• The algorithm effectively reduced artifacts and noise, enhancing image contour and grayscale distribution.

Conclusions:

• The proposed CBCT image reconstruction algorithm demonstrates potential for improving image quality.
• It can enhance diagnostic accuracy by producing clearer images from low and missing signal data.
• This advancement offers a valuable tool for medical imaging applications requiring high-fidelity reconstructions.