In this paper, we propose a detection method of low contrast structures in medical ultrasound images. Since noise speckle makes difficult the analysis of ultrasound images, two approaches based on the wavelet and Hermite-transforms for enhancement and noise reduction are compared. These techniques assume that speckle pattern is a random signal characterized by a Rayleigh distribution and affects the image as a multiplicative noise. For the wavelet-based approach, a Bayesian estimator at subband level for pixel classification is used. All the estimation parameters are calculated using an adjustment method derived from the first and second order statistical moments. The Hermite method computes a mask to find those pixels that are corrupted by speckle. In this work, we consider a statistical detection model that depends on the variable size and contrast of the image speckle. The algorithms have been evaluated using several real and synthetic ultrasound images. Combinations of the implemented methods can be helpful for automatic detection applications of tumors in mammographic ultrasound images. The employed filtering techniques are quantitatively and qualitatively compared with other previously published methods applied on ultrasound medical images.