Speckle Noise Reduction in Medical Ultrasound Images Using Particle Swarm Optimization with Artificial Neural Networks: Comparative Approach

SPECKLE NOISE REDUCTION IN MEDICAL ULTRASOUND IMAGES USING PARTICLE SWARM OPTIMIZATION WITH ARTIFICIAL NEURAL NETWORKS: COMPARATIVE APPROACH Er. Manpreet Kaur1 and Er. Danvir Mandal2 1CEC, Landran, Mohali E-mail: Manpreet.n.kaur @gmail.com 2IET BHADDAL, Ropar E-mail: [email protected] Medical images are normally affected by noise due to various sources of interferences and other phenomena that affect the process of measurement in an imaging and acquisition system. Speckle noise is a random mottling of the image with bright and dark spots, which destroys fine details , degrades the quality of image and detect ability of low-contrast lesions. Speckle noise occurrence is often undesirable, since it badly affects the tasks of human interpretation and diagnosis. Speckle removal is thus a critical pre-processing step in medical ultrasound image, so that the features of interest for diagnosis are not lost. In ultrasound images, the speckle energy is comparable to the signal energy in a wide range of frequency bands. Several speckle reduction techniques are applied to ultrasound images in order to reduce the noise level and improve the visual quality for better diagnoses. In the proposed method we use a three layer feed forward neural network and its optimization is done firstly by using back propagation method and then with particle swarm optimization to change the weights to achieve minimum mean square error in the image. Then the results of Back propagation and PSO are compared. By means of experimental results it has been shown that PSO yields far better results than the Back propagation algorithm. For the image quality performance measure we used mean square error (MSE) and Peak signal-to-noise ratio (PSNR) as the statistical parameters, as they are better measurements for speckle noise.