Efficient Approach for De-Speckling Medical Ultrasound Images Using Improved Adaptive Shock Filter

Abstract

The problem of filtering medical images is regarded one of the most important challenges that researchers are competing to solve it, where the filtered image helps to get the correct diagnosis of the diseases. This paper introduces an effective approach for filtering the medical ultrasound images. The main type of noise which corrupts the ultrasound images is the speckle noise. There are many methods for de-speckling this type of images addressed by the researchers including classical filters such as Weiner, Kuan, and Lee and adaptive filters such as shock filter. The performance of the proposed approach of this paper is compared with these filters using three performance evaluation metrics: "Peak Signal to Noise Ratio (PSNR)", "Mean Square Error (MSE)", and "Universal Image Quality Index (UIQ)". The empirical results illustrate that the proposed approach outperforms better than the others in term of these evaluation criteria. The proposed approach at noise variance=0.5 achieved the following values: (PSNR=32.0847db, MSE= 0.0962, and UIQ= 0.9829).