On the use of inverse scattering tomographic methods for quantitative ultrasound techniques

Abstract

The application of the distorted Born iterative method (DBIM) is proposed to construct tomographic images of small scale structures. Specifically, DBIM is proposed as a tool to construct better ultrasonic models of cells in order to improve the accuracy of quantitative ultrasound (QUS) analysis. An alternative implementation of DBIM valid for measurement configurations where the receiver positions are fixed is discussed. In addition, a method for the selection of the regularization parameter is determined, based on the Rayleigh quotient iteration. QUS imaging has been used to differentiate malignant from benign tissue. Recently, cells have been hypothesized as an important source of scattering in tissue. Generating tomographic images of sound speed and density of cells would allow better models to be constructed. Because the contrast mechanism in cells is unknown, the effect of variable density is explored. Stage II–III Xenopus oocytes are proposed as an initial cell model for scattering, with typical sizes ranging between 300 and 600 μm. Based on measurements of the propagation delay through oocytes, it was determined that their diameters correspond to at most 24 resolution cells when using single‐frequency DBIM. Hence, frequency hopping may be required to discriminate smaller features inside the oocytes.