An extended non-local means algorithm: Application to brain MRI

• Published in: International journal of imaging systems and technology Vol. 24; no. 4; pp. 293 - 305
• Main Authors: Iftikhar, Muhammad Aksam, Jalil, Abdul, Rathore, Saima, Ali, Ahmad, Hussain, Mutawarra
• Format: Journal Article
• Language: English
• Published: Hoboken, NJ Blackwell Publishing Ltd 01.12.2014; Wiley; Wiley Subscription Services, Inc
• Subjects: Algorithms; Applied sciences; Artificial intelligence; Biological and medical sciences; brain MRI; Computer science; control theory; systems; denoising; Exact sciences and technology; Investigative techniques, diagnostic techniques (general aspects); Medical sciences; Nervous system; nonlocal means; Pattern recognition. Digital image processing. Computational geometry; Radiodiagnosis. Nmr imagery. Nmr spectrometry; Rician noise; wavelet; Brain; Computer vision; Mixing; Subband decomposition; Noise reduction; brain MRI; Central nervous system; Subband; Nonlocal means; Nuclear magnetic resonance imaging; Adaptive method; Rician noise; Encephalon; Noise level; denoising; Wavelet transformation; Efficiency; wavelet; Rician distribution; Signal to noise ratio
• ISSN: 0899-9457; 1098-1098
• DOI: 10.1002/ima.22106

ABSTRACT Improved adaptive nonlocal means (IANLM) is a variant of classical nonlocal means (NLM) denoising method based on adaptation of its search window size. In this article, an extended nonlocal means (XNLM) algorithm is proposed by adapting IANLM to Rician noise in images obtained by magnetic resonance (MR) imaging modality. Moreover, for improved denoising, a wavelet coefficient mixing procedure is used in XNLM to mix wavelet sub‐bands of two IANLM‐filtered images, which are obtained using different parameters of IANLM. Finally, XNLM includes a novel parameter‐free pixel preselection procedure for improving computational efficiency of the algorithm. The proposed algorithm is validated on T1‐weighted, T2‐weighted and Proton Density (PD) weighted simulated brain MR images (MRI) at several noise levels. Optimal values of different parameters of XNLM are obtained for each type of MRI sequence, and different variants are investigated to reveal the benefits of different extensions presented in this work. The proposed XNLM algorithm outperforms several contemporary denoising algorithms on all the tested MRI sequences, and preserves important pathological information more effectively. Quantitative and visual results show that XNLM outperforms several existing denoising techniques, preserves important pathological information more effectively, and is computationallyefficient.