Ultrasound tissue characterizationby generalized GAMMA MRF model

摘要

In this paper we present a Markov Random Field (MRF) texture model for Radio-Frequency (RF) ultrasound data. The model is characterized by a set of parameters representing a powerful feature descriptor for the ultrasound envelope data. It combines the generalized gamma distribution (GGD) that describes global statistics of the envelope of backscattered echo with a local spatial interaction using MRF. We first construct the generalized gamma-MRF (GG-MRF) model and illustrate the role of its parameters. Applicability and performance are demonstrated on Choroidal melanoma.