Low-dose computed tomography image denoising using pixel level non-local self-similarity prior with non-local means for healthcare informatics.

Low-dose computed tomography (LDCT) has gained considerable attention for its ability to minimize patients' exposure to radiation thereby reducing the associated cancer risks. However, this reduction in radiation dose often results in degraded image quality due to the presence of noise and artifacts. To address this challenge, the present study proposes an LDCT image denoising method that leverages a pixel-level nonlocal self-similarity (NSS) prior in combination with a nonlocal means algorithm. The NSS prior identifies similar pixels within non-local regions, which proves more feasible and effective than patch-based similarity in enhancing denoising performance. By utilizing this pixel-level prior, the method accurately estimates noise levels and subsequently applies a non-local Haar transform to execute the denoising process. Furthermore, the study incorporates an enhanced version of a recently proposed nonlocal means algorithm. This revised approach uses discrete neighbourhood filtering properties to enable efficient, vectorized, and parallel computation on modern shared-memory platforms thereby reducing computational complexity. Experimental evaluations on publicly available benchmark dataset NIH-AAPM-Mayo Clinic Low-Dose CT Grand Challenge demonstrate that the proposed method effectively suppresses noise and artifacts while preserving critical image details. Both visual and quantitative comparisons confirm that this approach outperforms several state-of-the-art techniques in terms of image quality and denoising efficiency.