Sound Speed Images Research Articles

Inverse Scattering Image Reconstruction from Reflection and Transmission Data Observation with Fixed Transmitter/Receiver Pair Transducer

The inverse scattering imaging method based on reflection and transmission observation data, which was previously developed by the present author, is anticipated for its practical advantage in which the necessary range of observation view can be reduced to half compared to the conventional full observation view. In this report, we propose a modified method, which involves data observation carried out while rotating an object under the fixed condition of a transmitter/receiver pair, and a reflector. The inverse scattering reconstruction procedure is developed by applying the mirror-image theory. A special feature of the present method which is stressed is its superiority for the limitation of the observation view range much smaller than 180 deg. The validity of the method is tested using simulation and experimental data from the triangular specimen. It is shown that the quantitative precision of the reconstructed sound speed images is fairly good even when the limited viewing range is less than 120 deg.

Compensation of Linearization Approximation Error in Acoustic Inverse Scattering Problem: Quantitative Reconstruction of Attenuation and Sound Speed Images

For the realization of the acoustic parameter quantitative reconstruction of biological tissues by ultrasonic diffraction tomography, the author recently developed the backward propagation Rytov approximation to overcome the limitation of the weak scattering condition of tissuelike objects. A sound speed image with sufficient quantitative precision was realized. The problem, however, was that the precision of the attenuation image was limited when the sound speed contrast of the object was large. In this paper, a method for compensating the weak scattering approximation error is presented so that a compatible reconstruction of both sound speed and attenuation images can be realized. From numerical verification, the precision of the attenuation image after compensation is verified to be more superior than that before compensation.

Assessment of ultrasonic computed tomography in symptomatic breast patients by discriminant analysis

From 95 subjects imaged with both speed of sound and attenuation ultrasonic computed tomography (UCT), quantitative analyses are presented on 40 cases where unequivocal correlating clinical diagnoses are available. Using four attenuation and speed of sound parameters from different regions of interest in the breast, a linear discriminator detects cancer with approximately 90% sensitivity and specificity. Increased confidence in the predictive power of this small study is given by a modern test of predictive power (jackknifing) and by the fact that diagnostic discrimination remains as high as 85% when only two parameters are employed—attenuation and speed of sound in the lesion minus those values in the remaining central mammary tissues. Speed of sound images appear particularly useful in older, fatty breasts where pulse echo ultrasound is particularly lacking. While UCT in the form studied here is not likely to receive wide clinical acceptance in the near future, a combined UCT/pulse echo system might find wide clinical utility if it can be sufficiently convenient and inexpensive.

A Fundamental Evaluation of In Vivo Local Sound Speed Measurement by Crossed Beam Method (2nd report)

The possibility of in vivo sound speed mapping by the crossed beam method was studied by a calculation using the ray tracing technique and a fundamental experiment. The calculation result is that the sound speed image is shown as a very characteristic pattern, which gives useful information about in vivo sound speed distribution. It was confirmed that experimental values agreed with calculated values.

Median and spatial low-pass filtering in ultrasonic computed tomography.

To reduce noise- and reconstruction-related artifacts in ultrasound computed tomography, the use of combined median and Hamming-weighted spatial low-pass filtering is evaluated. The evaluation of the reconstruction filters uses both computer-generated projections of a known object with the least mean-square error criterion as well as a more subjective evaluation of conventional ultrasonic attenuation and speed of sound images.