Thursday, February 13, 1997
4:30-5:30 pm
1003 EECS
Abstract -
Ultrasonic tissue characterization techniques offer the promise of
non-invasive discrimination between normal and diseased tissue in
real-time by using existing imaging technology and appropriate digital
signal processing. Different parameters extracted from the processed
ultrasound echo data, such as integrated backscatter, attenuation,
reflection coefficients, scatterer distribution, and mean scatterer
spacing have been considered in the literature, with varying degrees
of success. Mean scatterer spacing (MSS) has been recognized as a tool
for tissue characterization for certain kinds of tissues and
diseases. Furthermore, it has recently been shown that MSS is a
valuable tool for non-invasive estimation of temperature change caused
by an externally applied heating field. Most of the MSS estimation
algorithms in the literature either are not robust in the presence of
irregularities in the scatterer distribution, or are computationally
demanding and cannot be used in real-time. We present a
computationally efficient and robust algorithm which uses the
magnitude and phase information of the ultrasound backscattered echo
spectra to estimate the MSS. This algorithm exploits the spectral
redundancy present in the echo signal by generating spectral lines
through a non-linear (quadratic) transformation of the RF echo signal.
Results of simulations comparing the performance of the proposed
algorithm and previous approaches from the literature are presented to
demonstrate the robustness of the proposed algorithm. Experiments
involving phantoms and in vitro tissue samples are also presented. The
feasibility of implementing a real-time MSS imaging system based on
the proposed method is discussed.
Biosketch -
Please refer to Mr. Simon's homepage
found through the link shown above.