The paper considers the application of statistical properties of backscattered ultrasonic signal for assessment of the trabecular bone status. Computer simulations were conducted to investigate the properties of the ultrasound pulse-echo signal, as it is received on the transducer surface after scattering in trabecular bone. The micro-architecture of trabecular bone was modeled by a random distribution of long and thin cylindrical scatterers of randomly varying diameters and mechanical properties, oriented perpendicular to the ultrasound beam axis. The received echo signal was calculated as a superposition of echoes from all the scatterers present in the scattering volume. The simulated signal envelope was used for statistical processing to compute various parameters like the mean amplitude, the amplitude MSR defined as the ratio of the mean to the standard deviation and the amplitude histogram. Results indicated that while for the well-defined trabeculae properties within the simulated bone structure the signal envelope values are Rayleigh distributed the significant departures from Rayleigh statistics may be expected as the thickness of trabeculae become random. The influence of the variation of mechanical properties of the bone tissue building the trabeculae on the bone backscattered signal parameters was not observed.
The paper presents a classification of the healthy skin and the skin lesions (basal cell carcinoma) basing on a statistics of the envelope of ultrasonic echoes. The echoes envelopes distributions were modeled using Rayleigh and K-distribution. The distributions were compared with empirical data to find which of them better models the statistics of the echo-signal obtained from the human skin. The results indicated that the K-distribution provides a better fit. Also, a characteristic parameter of the K-distribution, the effective number of scatterers (M), was investigated. The values of the M parameter, obtained for the skin cancer (basal cell carcinoma), were lower as compared to those obtained for the healthy skin. The results indicate that the statistical quantitative ultrasound parameters have a potential for extracting information useful for characterization of the skin condition.