Performance assessment of ultrasound kidney images using de-speckling algorithms

The key concern that occurs in non invasive Kidney stone diagnosis using Ultrasound (US) imaging is speckle noise, as it reduces the diagnostic quality of images, required for further medication. In this work, different preprocessing filters like Median, Adaptive median, Weiner and Wavelet domain filtering are applied to both normal and kidney stone US images, with results showcase, Neigh Sure Shrink is preeminent for kidney stone US images. Objective feature assessment of the different preprocessing filters are evaluated for noise variance from 0.01 and 0.08 along with level of decomposition in Neigh Sure Shrink against the parameters like Signal to Noise Ratio (SNR), Peak Signal to Noise Ratio (PSNR), Root Mean Square error (RMSE), Mean Squared Error (MAE) with the investigational outcome shows the preeminence of projected filters over the existing methods. Further, the segmentation method allows bifurcation of the US image to detect the kidney stone on basis of Region of Interest, followed by determining the presence of Centroid and area for kidney stone US images. The Graphical User Interface allows easiness in locating the area of kidney stone for kidney US images. � 2020, Engg Journals Publications. All rights reserved.