Tissue differentiation based on radiofrequency echographic signal local spectral content

An echographic method for investigating biological tissue internal structure in order to differentiate pathological regions is proposed in this work. The method, named RULES (radiofrequency ultrasonic local estimator), is based on interactive radiofrequency (RF) echographic signal processing. It permits to extract spectral parameters related to the actual organization and mechanical properties of investigated tissue. Spectral images are produced through a processing procedure, based on a statistical analysis of local estimators (LE). They are calculated from the spectral coefficients of discrete wavelet packet transform. The tissue portions, which exhibit the same LE homogeneous distribution, were put in correspondence with condition and type of tissue identified by histological studies. Results obtained on prostate pathology detection, breast tumor detection and carotid vessel plaque localization and differentiation are presented.