Quantitative soft-tissue characterization in human organs using texture/attenuation models

Summary form only given. Soft tissues in human organs (such as the liver and the prostate) have been characterized using ultrasound data. Using a combination of texture/attention modeling of RF data (obtained from an ultrasound scanner equipped with a quadrature detector), an approach to quantifying the different pathologies of the human tissue has been developed. The model is derived from the physics of the tissue-ultrasound interactions. It is based on modeling the homogeneous speckle field produced by soft-tissue scattering by a complex, circularly symmetric Gaussian Markov random field (CGMRF). In contrast to all prior attempts to quantify the speckle field, which were limited to the use of partial information such as first-order statistics, or the second-order statistics, and autocorrelation function, the CGMRF completely characterizes the speckle field