A constructive non-local means algorithm for low-dose computed tomography denoising with morphological residual...

Area of Science:

• Medical Imaging
• Image Processing
• Radiology

Background:

• Low-dose computed tomography (LDCT) reduces radiation exposure but introduces noise and artifacts.
• Image degradation in LDCT hinders accurate medical disease diagnosis.
• Existing denoising methods struggle to balance noise reduction with detail preservation.

Purpose of the Study:

• To develop an effective low-dose computed tomography image denoising algorithm.
• To address the challenges of noise and artifacts in LDCT imaging.
• To improve diagnostic performance by enhancing LDCT image quality.

Main Methods:

• A constructive non-local means algorithm was modified for LDCT denoising.
• The algorithm incorporates morphological residual processing for edge preservation.
• Vectorized and parallel implementation enables efficient computation on modern hardware.

Main Results:

• The proposed algorithm effectively reduces noise and artifacts in LDCT images.
• It preserves more textural and structural features compared to existing methods.
• Experimental results show significant improvements in image quality, both qualitatively and quantitatively.

Conclusions:

• The developed algorithm offers a competent solution for LDCT image denoising.
• It enhances edge preservation and overall image quality.
• This approach holds promise for improving diagnostic accuracy in low-dose CT imaging.