Gabor-based anisotropic diffusion for speckle noise reduction in medical ultrasonography

Abstract

In ultrasound (US), optical coherence tomography, synthetic aperture radar, and other coherent imaging systems, images are corrupted by multiplicative speckle noise that obscures image interpretation. An anisotropic diffusion (AD) method based on the Gabor transform, named Gabor-based anisotropic diffusion (GAD), is presented to suppress speckle in medical ultrasonography. First, an edge detector using the Gabor transform is proposed to capture directionality of tissue edges and discriminate edges from noise. Then the edge detector is embedded into the partial differential equation of AD to guide the diffusion process and iteratively denoise images. To enhance GAD's adaptability, parameters controlling diffusion are determined from a fully formed speckle region that is automatically detected. We evaluate the GAD on synthetic US images simulated with three models and clinical images acquired in vivo. Compared with seven existing speckle reduction methods, the GAD is superior to other methods in terms of noise reduction and detail preservation.

Segmentation of Lymph Nodes in Ultrasound Images Using U-Net Convolutional Neural Networks and Gabor-Based Anisotropic Diffusion

“…In this section, we introduce the GAD to suppress speckle and enhance nodal edges in the medical ultrasonography of LNs. The GAD is a speckle reduction method for denoising ultrasound images by employing a new edge detector based on the Gabor transform into the anisotropic diffusion [18]. If an input image is denoted as I(x, y), its Gabor transform is the convolution of I(x, y) with a family of Gabor kernels:…”

“…Here, only the imaginary part of the Gabor kernel is utilized for convolution [18]. An edge detector based on the Gabor transform, called the Gabor-based edge detector, is hence given by:…”

“…The Gabor-based anisotropic diffusion (GAD) captures the edge directionality with the Gabor-based edge detector. The GAD not only suppresses speckle noise in ultrasound but also preserves and enhances tissue edges, structures, and details [18]. Thus, it has the potential to be explored for noise reduction and edge enhancement so as to more accurately segment LNs in ultrasound.…”

Intelligent speckle reducing anisotropic diffusion algorithm for automated 3-D ultrasound images

“…In [8,21,22], the mean absolute error (MAE) between two successive diffusion steps is used to stop diffusion automatically, which is written as…”

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

“…Few years later Zhan et al modified non local means filter and added weight refining for despeckling of ultrasound images [14]. In the same year Zhang et al proposed Gabor based anisotropic diffusion for speckle noise reduction in medical ultrasonography image [15].…”