Adaptive spectral estimation of the attenuation in skin

In the field of biological tissue characterization, fundamental acoustic attenuation properties have demonstrated to have diagnostic importance. A nonstationary autoregressive (AR) model is employed to estimate the acoustical attenuation which is known to be dependent on frequency. A particular frequency such as maximum energy frequency is used to evaluate the spectral downward shift. Backscattered signals by a homogeneous medium of scatterers are supplied by a computer model with attenuation values ranging from 1 to 5 dB/cm MHz and 45 MHz transducer central frequency. The proposed adaptive AR algorithm is evaluated with these experimental conditions. Preliminary obtained results promise encouraging perspectives for in vitro and in vivo applications, namely in the skin identification area.