(A) new smoothing algorithm of the ultrasound attenuation estimation using the LMSE filter and median filter

Abstract

The ultrasound attenuation coefficient is known to increase with frequency for soft biological tissue within the diagnostic frequency range. The frequency dependent attenuation coefficient becomes an important parameter in the field of clinical tissue characterization, especially in liver, and the ultrasonic quantitative B-scan imaging. But, because of the low S/N ratio, the estimator variance becomes very large. Henceforth, it has been a difficult problem to estimate the attenuation slope from the backscattered ultrasound signal. This thesis suggests a new ultrasound attenuation estimator based on the calibrated log spectral difference. The proposed estimation process consists of Wiener filtering, Kalman filtering and median filtering. The simulation results and experiment on tissue equivalent(TE) phantom shows that combination of the filterings has a better performance with a considerably reduced estimator variance than the conventional estimator.