Speckle suppression for 3D ultrasound images using nonlinear multiscale wavelet diffusion

Abstract

We introduce a new speckle suppression approach for 3-D ultrasound images. The proposed method seeks to enhance volume visualization of 3-D ultrasound images, and improve the accuracy of volume determination. Extended from 2-D nonlinear multiscale wavelet diffusion, the proposed method is developed on the basis of an integration of the 3-D nonlinear diffusion and 3-D dyadic wavelet transform techniques. Our approach uses the normalized wavelet modulus as an edge map to expose the intrinsic speckle/edge relation. Relying on the statistical analysis of this edge map, the method is able to classify homogenous speckle regions and edges in a 3-D volume, as well as provide strong speckle suppression and boundary preservation. Our method is validated using both synthetic and real 3-D ultrasonic images. Performance improvement over other filters is quantified by evaluating standard quality indices.