A Hybrid Approach for Image Denoising in Ultrasound Carotid Artery Images

Abstract

Reliable, effective, and exact estimations of the geometry of the Common Carotid Artery (CCA) are necessary to identify stroke and heart attack prior. The utilization of ultrasound imaging in medicinal diagnosis is well recognized. Denoising speckle in ultrasound is vital. This paper gives thresholding of curvelet for denoising of carotid artery ultrasound images. The hybrid filter preserves radiometric information, edge information, and spatial resolution. Curvelet transform combines speckle reducing and edge preserving properties of the image because it exploits the instantaneous coefficient of variation which is a function of local gradient magnitude and Laplacian operators. Blurred thin edges and low contrast fine features are preserved by thresholding. The filtering methods performance is valuated in terms of eight performance metrics. The proposed filter is compared with existing filters which were used for the carotid artery ultrasound application.