Peter Huthwaite

High-resolution imaging without iteration: A fast and robust method for breast ultrasound tomography

Breast ultrasound tomography has the potential to improve the cost, safety, and reliability of breast cancer screening and diagnosis over the gold-standard of mammography. Vital to achieving this potential is the development of imaging algorithms to unravel the complex anatomy of the breast and its mechanical properties. The solution most commonly relied upon is time-of-flight tomography, but this exhibits low resolution due to the presence of diffraction effects. Iterative full-wave inversion methods present one solution to achieve higher resolution, but these are slow and are not guaranteed to converge to the correct solution. Presented here is HARBUT, the hybrid algorithm for robust breast ultrasound tomography, which utilizes the complementary strengths of time-of-flight and diffraction tomography resulting in a direct, fast, robust and accurate high resolution method of reconstructing the sound speed through the breast. The algorithm is shown to produce accurate reconstructions with realistic data from a complex three-dimensional simulation, with masses as small as 4 mm being clearly visible.

Mode selection for corrosion detection in pipes and vessels via guided wave tomography

Guided wave tomography offers a method to accurately quantify wall thickness losses in pipes and vessels caused by corrosion, using ultrasonic waves transmitted over distances of approximately 1 to 2 m, and measured by an array of transducers. These measurements are then used to reconstruct a map of wall thickness throughout the inspected region. To achieve accurate estimations of remnant wall thickness, it is vital that a suitable Lamb mode is chosen. This paper presents a detailed evaluation of the two most suitable modes, S0 and A0, to compare their performance using both numerical and experimental data. The sensitivity of A0 to thickness variations was shown to be superior to S0; however, the attenuation from A0 when a liquid loading was present was much higher than S0. A0 was less sensitive to the presence of coatings on the surface than was S0. Finally, it was shown that both modes could achieve a similar level of resolution in the plane of the plate surface.

Numerical design optimization of an EMAT for A0 Lamb wave generation in steel plates

An electromagnetic acoustic transducer (EMAT) for A0 Lamb wave generation on steel plates is developed to operate at 0.50 MHz-mm. A key objective of the development is to maximize the excitation and reception of the A0 mode, while minimizing those of the S0 mode. The chosen EMAT design consists of an induction coil and a permanent magnet. A finite element (FE) model of the EMAT is developed, coupling the electromagnetic and elastodynamic phenomena. An optimization process using a genetic algorithm is implemented, employing the magnet diameter and liftoff distance from the plate as design parameters and using the FE model to calculate the fitness. The optimal design suggested by the optimization process is physically implemented and the experimental measurements are compared to the FE simulation results. In a further step, the variations of the design parameters are studied numerically and the proposed EMAT design exhibits a robust behavior to small changes of the design parameters.

Finite element modelling of elastic wave scattering within a polycrystalline material in two and three dimensions

Finite element modelling is a promising tool for further progressing the development of ultrasonic non-destructive evaluation of polycrystalline materials. Yet its widespread adoption has been held back due to a high computational cost, which has restricted current works to relatively small models and to two dimensions. However, the emergence of sufficiently powerful computing, such as highly efficient solutions on graphics processors, is enabling a step improvement in possibilities. This article aims to realise those capabilities to simulate ultrasonic scattering of longitudinal waves in an equiaxed polycrystalline material in both two (2D) and three dimensions (3D). The modelling relies on an established Voronoi approach to randomly generate a representative grain morphology. It is shown that both 2D and 3D numerical data show good agreement across a range of scattering regimes in comparison to well-established theoretical predictions for attenuation and phase velocity. In addition, 2D parametric studies illustrate the mesh sampling requirements for two different types of mesh to ensure modelling accuracy and present useful guidelines for future works. Modelling limitations are also shown. It is found that 2D models reduce the scattering mechanism in the Rayleigh regime.

A new regularization technique for limited-view sound-speed imaging

Reconstructing sound-speed maps from the limited view offered by a linear array of ultrasonic sensors has been a long-standing challenge in medical diagnostics and nondestructive evaluation. Because of the limited range of angles that can be used to interrogate the volume beneath the array, the inverse problem of retrieving sound-speed maps from scattering measurements is highly ill-posed. The missing angles cause significant artifacts that degrade the image by altering the values of sound speed and producing ghost features. This paper introduces the virtual image space component iterative technique (VISCIT), which addresses the limited-view problem by introducing a new regularization technique which iteratively compensates for the missing components by applying an adaptive threshold to the reconstruction. The effectiveness of the method in yielding high-accuracy sound-speed maps is demonstrated using a complex numerical phantom and validated experimentally with an agar phantom. It is shown that sound-speed contrast as low as 1.3% is readily detectable, thus paving the way for more sensitive and selective detection of damage precursors and early stage diseases.

Experimental Studies of the Inspection of Areas With Restricted Access Using A0 Lamb Wave Tomography

Corrosion damage in inaccessible regions presents a significant challenge to the petrochemical industry, and determining the remaining wall thickness is important to establish the remaining service life. Guided wave tomography is one solution to this and involves transmitting Lamb waves through the area of interest and, subsequently, using the received signals to reconstruct a thickness map of the remaining wall thickness. This avoids the need to access all points on the surface, making the technique well suited to inspection for areas with restricted access. The influence of these areas onto the ability to detect and size surface conditions, such as corrosion damage, using guided wave tomography is assessed. For that, a guided wave tomography system is employed, which is based on low-frequency A0 Lamb waves that are excited and detected with two arrays of electromagnetic acoustic transducers. Two different defect depths are considered with different contrasts relative to the nominal wall thickness, both of which are smoothly varying and well-defined. The influence of areas with restricted surface access, support locations, pipe clamps, and STOPAQ(R) coatings is experimentally tested, and their influence assessed through comparison to a baseline reconstruction without the respective restriction in place, demonstrating only a small influence on the detected value of the remaining wall thickness.

Robust helical path separation for thickness mapping of pipes by guided wave tomography

Pipe wall loss caused by corrosion can be quantified across an area by transmitting guided Lamb waves through the region and measuring the resulting signals. Typically the dispersive relationship for these waves, which means that wave velocity is a known function of thickness, is exploited, enabling the wall thickness to be determined from a velocity reconstruction. The accuracy and quality of this reconstruction is commonly limited by the angle of view available from the transducer arrays. These arrays are often attached as a pair of ring arrays on either side of the inspected region, and due to the cylindrical nature of the pipe, waves are able to travel in an infinite number of helical paths between any two transducers. The first arrivals can be separated relatively easily by time gating, but by using just these components the angle of view is severely restricted. To improve the viewing angle, it is necessary to separate the wavepackets. This paper provides an outline of a separation approach: initially the waves are backpropagated to their source to align the different signals, then a filtering technique is applied to select the desired components. The technique is applied to experimental data and demonstrated to robustly separate the signals.

ON THE CONVERGENCE OF FINITE ELEMENT SCATTERING SIMULATIONS

The ability to produce accurate scattering data for imaging is important in the field of NDT for both the development of new algorithms and iterative methods which improve the scatterer reconstruction by trial and error. Error analysis for FE simulations is typically performed for plane wave propagation in a homogeneous space. In this paper it is demonstrated that the mesh refinement must be significantly improved to achieve the same accuracy for scattering problems. A study of meshing techniques suggests that a uniform mesh is necessary to obtain a suitable dynamic range.

Improving accuracy through density correction in guided wave tomography

The accurate quantification of wall loss caused by corrosion is critical to the reliable life estimation of pipes and pressure vessels. Traditional thickness gauging by scanning a probe is slow and requires access to all points on the surface; this is impractical in many cases as corrosion often occurs where access is restricted, such as beneath supports where water collects. Guided wave tomography presents a solution to this; by transmitting guided waves through the region of interest and exploiting their dispersive nature, it is possible to build up a map of thickness. While the best results have been seen when using the fundamental modes A0 and S0 at low frequency, the complex scattering of the waves causes errors within the reconstruction. It is demonstrated that these lead to an underestimate in wall loss for A0 but an overestimate for S0. Further analysis showed that this error was related to density variation, which was proportional to thickness. It was demonstrated how this could be corrected for in the reconstructions, in many cases resulting in the near-elimination of the error across a range of defects, and greatly improving the accuracy of life estimates from guided wave tomography.

Detecting breast microcalcifications using super-resolution and wave-equation ultrasound imaging: a numerical phantom study

Ultrasound image resolution and quality need to be significantly improved for breast microcalcification detection. Super-resolution imaging with the factorization method has recently been developed as a promising tool to break through the resolution limit of conventional imaging. In addition, wave-equation reflection imaging has become an effective method to reduce image speckles by properly handling ultrasound scattering/diffraction from breast heterogeneities during image reconstruction. We explore the capabilities of a novel super-resolution ultrasound imaging method and a wave-equation reflection imaging scheme for detecting breast microcalcifications. Super-resolution imaging uses the singular value decomposition and a factorization scheme to achieve an image resolution that is not possible for conventional ultrasound imaging. Wave-equation reflection imaging employs a solution to the acoustic-wave equation in heterogeneous media to backpropagate ultrasound scattering/diffraction waves to scatters and reconstruct images of heterogeneities. We construct numerical breast phantoms using in vivo breast images, and use a finite-difference wave-equation scheme to generate ultrasound data scattered from inclusions that mimic microcalcifications. We demonstrate that microcalcifications can be detected at full spatial resolution using the super-resolution ultrasound imaging and wave-equation reflection imaging methods.