Wavelet based non local means filter for despeckling of intravascular ultrasound image

Abstract

In this paper, a novel methodology has been proposed for speckle noise reduction of intravascular ultrasound (IVUS). IVUS, a standard coronary artery diagnosis imaging protocol, is degraded with speckle noise due to the coherent interferences of the ultrasound reflected from the scatters. Presence of such noise creates problem in segmentation and classification of these images. Non local means filter has been applied in wavelet domain to smooth out noise interferences to improve the visual characteristics of IVUS insight. Eventually, a comparison study has also been performed with various filters viz. anisotropic diffusion filter, nonlinear median filter and geometric nonlinear diffusion filter. Mean squared error (MSE) and peak signal to noise ratio (PSNR) has been used to evaluate the efficiency of the proposed filtering methodology.

Performance analysis of various denoising filters on intravascular ultrasound coronary artery images

“…Wavelets are being used for speckle reduction in many medical imaging applications. The effects of wavelets in clinical imaging have proven to be effective and used by many researchers 1,59 . The noise‐free image could be obtained using thresholding (either hard or soft thresholding).…”

“…The hard‐thresholding function eliminates wavelet coefficients under a certain threshold. It is given by Equation (): where, W represents wavelet coefficient; is the threshold, and is the processed wavelet coefficients by hard‐ thresholding 1,31 …”

“…This technique is beneficial than a normal anisotropic diffusion filter where only region‐based comparison is done to calculate the weighted average of the target pixel and no pixel‐based information is used. For calculating the NL‐means, consider a gray‐scale noisy image given in Equation (): The value for NL [ y ]( x ), is calculated using the weighted average of pixels in the image as given in Equation (), where x denotes the pixel value 1,21 where, {w ( x,y ) } y denotes the similarity between the pixels x and y , that meet the conditions such as 0 ≤ w ( x,y ) ≤ 1 and ∑ y w ( x,y ) = 1 .…”

“…Intravascular ultrasound (IVUS) is a catheter‐based technology that produces cross‐sectional images of the coronary artery to effectively perform the quantitative and qualitative assessment of atherosclerotic disease. In an IVUS examination, the extent of plaque deposition and arterial borders can be easily monitored and analyzed, which is not possible with other imaging technologies 1 . The speckle is an unavoidable feature of ultrasonic images that messes up the capacity of medical imaging.…”

Classification of Intravascular Ultrasound Images Based on Non-Negative Matrix Factorization Features and Maximum Likelihood Classifier

“…IVUS image filtering using wavelet and non local means filter is discussed in [9]. Initially the IVUS images are given to wavelet transform that produce approximate and detailed coefficients.…”