Ron Roberts

Computational Study of Grain Scattering Effects in Ultrasonic Measurements

A two‐dimensional scalar model has been established to study the scattering caused by an inhomogeneous medium with granular microstructure. Time domain signals are calculated using a 64‐node PC cluster. Special test problems are set up to compare the effects of single and multiple scattering and to quantify errors introduced by the Born approximation. Computed results for backscattered RMS grain noise results are compatible with recent experimental measurements.

Therapy planning and monitoring of tissue ablation by high intensity focused ultrasound (HIFU) using imaging and simulation

High intensity focused ultrasound (HIFU) “cooks” or ablates the target tissue at the focus of the ultrasound beam by thermal and cavitation effects. The HIFU is emerging as a non-invasive method for tumor ablation. The HIFU application for tissue ablation requires tools for dosimetry therapy planning, and real-time feedback of the intended and actual target tissues. Pretreatment planning is an important step for a successful HIFU therapy outcome. Typically, the therapy planning approach involves the use of pretreatment imaging data, defining the target and surrounding tissues by manual or semiautomatic segmentation, development of a 3-D anatomy model of the region of interest from segmentation or registration with a reference dataset, simulation of the HIFU beam and thermal dosimetry around the target tissue, display and 3-D visualization of imaging and simulation data, and review of the treatment plan options. Recent developments in therapy planning using imaging are targeted for specific applications such as prostate cancer using 3-D ultrasound images and uterine fibroids using MRI. However, significant developments have been accomplished in image guidance and feedback during the delivery of HIFU treatments. This talk reviews recent work towards therapy planning and presents approaches for developing strategies for HIFU therapy. It describes general and target-specific techniques and software tools for HIFU treatment planning using pretherapy imaging, and monitoring and controlling the HIFU delivery and tissue lesion using 1D, 2D and 3D ultrasound imaging. This aids development of optimized, high-precision HIFU applications for a controlled ablation of the target tumor. It also potentially reduces the overall treatment duration and exposure to non-target tissues.

Ground penetrating radar applied to rebar corrosion inspection

In this paper we investigate the use of ground penetrating radar (GPR) to detect corrosion-induced thinning of rebar in concrete bridge structures. We consider a simple pulse/echo amplitude-based inspection, positing that the backscattered response from a thinned rebar will be smaller than the similar response from a fully-intact rebar. Using a commercial 1600-MHz GPR system we demonstrate that, for laboratory specimens, backscattered amplitude measurements can detect a thinning loss of 50% in rebar diameter over a short length. GPR inspections on a highway bridge then identify several rebar with unexpectedly low amplitudes, possibly signaling thinning. To field a practical amplitude-based system for detecting thinned rebar, one must be able to quantify and assess the many factors that can potentially contribute to GPR signal amplitude variations. These include variability arising from the rebar itself (e.g., thinning) and from other factors (concrete properties, antenna orientation and liftoff, etc.). We...

Influence of forward scattering on ultrasonic attenuation measurement

A two dimensional scalar model has been established to study the scattering phenomenon caused by an inhomogeneous medium with granular microstructure. The scattering problem has been formulated as a volume integral, which can be solved by using a FFT-based Neumann iteration. This model has been used to simulate the different apparent attenuation values under different experiment setups and the simulation results are presented and discussed in this report

Study of the effect of microstructure on ultrasonic signal attenuation

The apparent attenuation arising from material microstructure is studied using a scalar 2-D model. The model simulates ultrasonic beam distortion during the propagation through a weakly scattering inhomogeneous microstructure that is defined by specifying the sonic velocity at each point in space. Uniform sized rectangular cells are used to simulate the elongated macrograins in Ti alloys. Numerical calculations are performed for both focused and planar transducers. Results are compared with measured attenuation results for engine titanium alloys.

An Interactive HIFU Therapy Planning Using Simulation & Visualization

Planning of HIFU therapy using patient‐specific data is crucial for successful results. We are developing integrated tools for HIFU therapy planning using pre‐treatment imaging, HIFU simulation, and interactive visualization. We employ techniques for 2D/3D image segmentation to generate layered surface model; ultrasound simulation of beam intensity distribution and HIFU temperature profiles; visualization of integrated 3D anatomy and US beam simulation; and interactive strategy planning for studying the approaches of HIFU delivery. We present a systematic approach for developing and integrating these tools in interactive software using C++, OpenGL, C♯.Net, and ITK. With interactive visualization of the HIFU simulation parameters and the tissue effects, a range of scenario could be studied for an optimal HIFU delivery for a given patient. This continuing work has potential to aid development of optimized high‐precision HIFU dosimetry and patient‐specific planning strategies by reducing the guess work on do...

A 2-D Numerical Simulation Study of Microstructure-Induced Ultrasonic Beam Distortions

In past work we have introduced an algorithm for computing ultrasonic pressure fields propagating through inhomogeneous media. The algorithm uses a volume integral formulation of the scattering problem, which is solved by an efficient FFT‐based Neumann iteration. In inspections of Titanium‐alloy, jet‐engine materials microstructure‐induced ultrasonic beam distortions and associated ultrasonic signal fluctuations are often seen. Our computational approach can be used to study these phenomena. In the present work, 2‐D numerical simulations have been carried out to study the statistics governing ultrasonic beam distortions. Several microstructures are considered, each of which is modeled as a collection of uniform‐sized rectangular cells with different sound speeds. The rectangular cells simulate macrograins in jet‐engine alloys. Sound beams computed for propagation through the inhomogeneous microstructures are compared to those for the homogeneous case. From the differences we obtain statistical description...

Optimization of Transmission Field for DDF‐Based Phased‐Array Inspection

Phased array ultrasonic inspections are being used on an ever‐wider scale in industrial applications. One of the phased array capabilities is a dynamic‐depth focusing (DDF) allowing an effective extension of focal zone. The use of DDF may potentially lead to decreased inspection times and costs for manufactures while maintaining a required sensitivity. The DDF is based on a dynamic re‐focusing on reception resulting in a relatively uniform apparent acoustic field through a focusing range. A further improvement may be achieved by optimizing an acoustic field excited on transmission. This paper discusses a transmission field optimization aimed at creating a uniform apparent field and shows corresponding experimental results. A simple visualization tool for qualitative optimization of the transmission field is also discussed.

Two‐Step Monte‐Carlo Simulation of Ti Alloy Microstructures for Studies of Ultrasonic Beam Fluctuations

In past work, algorithms have been developed to compute the manner in which a sonic beam becomes distorted when passing through a two‐dimensional inhomogeneous metal microstructure. We now take up the problem of generating realistic model microstructures to serve as inputs for the computations. In particular, we consider microstructures similar to those found in certain jet‐engine titanium alloys where, during cooling, colonies of alpha‐phase (hexagonal) micrograins develop from higher‐temperature beta‐phase (cubic) macrograins. Here, model microstructures are generated using a 2‐step Monte Carlo method based on the Potts model for grain growth. In the first step large, equiaxed beta macrograins are generated having randomly‐oriented principal axes. In the second step, micrograins are grown inside the macrograin boundaries, taking into account the 6 possible orientation variants a micrograin can have with respect to its parental macrograin. Stretching of grains, either in step 1 or step 2, can be used to produce elongated microstructures like those seen in practice. Longitudinal wave speeds for the alpha micrograins are calculated from their spatial orientations by solving the Christoffel equation. The simulated microstructures are then passed to the numerical wave‐propagation algorithm to study wave‐microstructure interactions. Model predictions for P/E back‐wall responses and backscattered noise are presented.In past work, algorithms have been developed to compute the manner in which a sonic beam becomes distorted when passing through a two‐dimensional inhomogeneous metal microstructure. We now take up the problem of generating realistic model microstructures to serve as inputs for the computations. In particular, we consider microstructures similar to those found in certain jet‐engine titanium alloys where, during cooling, colonies of alpha‐phase (hexagonal) micrograins develop from higher‐temperature beta‐phase (cubic) macrograins. Here, model microstructures are generated using a 2‐step Monte Carlo method based on the Potts model for grain growth. In the first step large, equiaxed beta macrograins are generated having randomly‐oriented principal axes. In the second step, micrograins are grown inside the macrograin boundaries, taking into account the 6 possible orientation variants a micrograin can have with respect to its parental macrograin. Stretching of grains, either in step 1 or step 2, can be used to ...

LEAK LOCATION IN SPACECRAFT SKIN WITH ULTRASONIC ARRAYS

An array‐based leak location sensor has been developed and improved for application onboard spacecraft. Introduction of a diced piezoelectric element has permitted the successful utilization of lower frequency bands to perform the location. With this new device location across five integral stiffeners has been demonstrated, despite the stiffeners strong frequency‐dependent scattering. Spatial Fourier transforms of plate wave signals determine leak location independent of the effects of dispersive multiple mode signal transport. The performance of the diced‐element array sensor on a 2×2‐m integrally stiffened aluminum plate is demonstrated.