Use of Nakagami distribution and logarithmic compression in ultrasonic tissue characterization

Our previous simulation study showed that the Nakagami parameter estimated using ultrasonic backscattered envelopes compressed by logarithmic computation, denoted by m_log, is more sensitive than the original Nakagami parameter m calculated using uncompressed envelopes for detecting the variation of scatterer concentration in tissues. This study made measurements on phantoms in order to further verify the performance of m_log in quantifying the properties of biological tissues. The ultrasonic backscattered signals from phantoms with different scatterer concentrations were acquired using 5 MHz focused and non-focused transducers. The Hilbert transform and logarithmic compression were in turn applied to the backscattered signals to obtain the uncompressed and compressed envelopes for estimating m and m_log. The experimental results showed that, for both focused and non-focused transducers, the m_log parameter is indeed more sensitive than the m parameter in differentiating various scatterer concentrations. This may assist in the classification of scatterer properties using the Nakagami statistical model.