A new adaptive denoising method based on empirical mode decomposition in ultrasonic elastography

Important soft tissue parameters such as elasticity can be obtained by ultrasonic elastic imaging. Nonetheless, the radio frequency (RF) data received by ultrasonic transducer are always mixed up with noises. In this paper, we develop a new adaptive signal processing method for ultrasonic elastography. The key part of the method is the empirical mode decomposition (EMD) which a method associated with the Hilbert-Huang transform (HHT). EMD method is well suited to analyzing non-stationary and non-linear signal such as biomedical signal. The raw RF data is firstly decomposed into several intrinsic mode functions (IMF) using EMD. Then we analyze its energy distribution and set a to control the signal reconstruction. In the experimental study, the data gathered both from tissue mimicking phantom and volunteer are tested. The results obtained with the new method are in good agreement with those obtained with the conventional one, the interferences are removed effectively especially in the low frequency domain.