Adaptive Artifact Suppression in Ultrasound Image

The aim of this report is to present artifact suppression algorithms for reducing the reverberation artifacts in ultrasonic imaging. These reverberation patterns always occurred as signals among specular reflectors. In ultrasound applications, we need to reduce the artifacts but at the same time to preserve other useful information. Because of the periodic pattern of the artifacts, we will apply a 2D FFT in un-overlap blocks of the image to extract and remove the peaks which are corresponding to artifacts in the frequency domain. The 2D FFT can provide the information in one more dimension compared with the 1D FFT from which we could differentiate the real artifacts and some useful information, for example, the vessel boundary. Test results from several in vivo images show that this new technique can suppress artifact effectively and preserve useful information at the same time.