We analyze the principles underlying minimum variance distortionless response (MVDR) beamforming in order to integrate it into a pixel-based algorithm. There is a challenge posed by the low echo signal-to-noise ratio (eSNR) when calculating beamformer contributions at pixels far away from the beam centreline. Together with the well-known scarcity of samples for covariance matrix estimation, this reduces the beamformer performance and degrades the image quality. To address this challenge, we implement the MVDR algorithm in two different ways. First, we develop the conventional minimum variance pixel-based (MVPB) beamformer that performs the MVDR after the pixel-based superposition step. This involves a combination of methods in the literature, extended over multiple transmits to increase the eSNR. Then we propose the coherent MVPB beamformer, where the MVDR is applied to data within individual transmits. Based on pressure field analysis, we develop new algorithms to improve the data alignment and matrix estimation, and hence overcome the low-eSNR issue. The methods are demonstrated on data acquired with an ultrasound open platform. The results show the coherent MVPB beamformer substantially outperforms the conventional MVPB in a series of experiments, including phantom and in vivo studies. Compared to the unified pixel-based beamformer, the newest delay-and-sum algorithm in [1], the coherent MVPB performs well on regions that conform to the diffuse scattering assumptions on which the minimum variance principles are based. It produces less good results for parts of the image that are dominated by specular reflections.

This paper describes the development and evaluation of a new beamforming strategy based on pixel-based focusing for ultrasound linear array systems. We first implement conventional pixel-based beamforming in which the transmitted wave is assumed as spherical and diverging from the centre of the transmit subaperture. This assumed wave-shape is only valid within a limited angle on each side of the beam and this restricts the number of different subaperture positions from which data can be combined to improve image quality. By analyzing the field patterns, we propose a new unified pixel-based beamforming algorithm that better adapts to the non-spherical wave-shape of the transmit beam. This approach enables us to select the best-possible signal from each transducer waveform for data superposition. In simulations and a phantom study, we show that the unified pixel-based beamformer offers significant improvements in image quality compared to other delay-and-sum methods but at a higher computational cost. The new algorithm also demonstrates robust performance in a limited in vivo study. Overall, the results show that it is potentially of value in clinical applications.

Displacement estimation is a key step in the evaluation of tissue elasticity by quasistatic strain imaging. An efficient approach may incorporate a tracking strategy whereby each estimate is initially obtained from its neighbours' displacements and then refined through a localized search. This increases the accuracy and reduces the computational expense compared with exhaustive search. However, simple tracking strategies fail when the target displacement map exhibits complex structure. For example, there may be discontinuities and regions of indeterminate displacement caused by decorrelation between the pre- and post-deformation radio frequency (RF) echo signals. This paper introduces a novel displacement tracking algorithm, with a search strategy guided by a data quality indicator. Comparisons with existing methods show that the proposed algorithm is more robust when the displacement distribution is challenging.

A freehand scanning protocol is the only way to acquire arbitrary large volumes of three-dimensional ultrasound (US) data. For some applications, multiple freehand sweeps are required to cover the area of interest. Aligning these multiple sweeps is difficult, typically requiring nonrigid image-based registration as well as the readings from the spatial locator attached to the US probe. Conventionally, nonrigid warps are achieved through general elastic spline deformations, which are expensive to compute and difficult to constrain. This paper presents an alternative registration technique, where the warp's degrees of freedom are carefully linked to the mechanics of the freehand scanning process. The technique is assessed through an extensive series of in vivo experiments, which reveal a registration precision of a few pixels with comparatively little computational load.

The ability to estimate a surface from a set of cross sections allows calculation of the enclosed volume and the display of the surface in three-dimensions. This process has increasingly been used to derive useful information from medical data. However, extracting the cross sections (segmenting) can be very difficult, and automatic segmentation methods are not sufficiently robust to handle all situations. Hence, it is an advantage if the surface reconstruction algorithm can work effectively on a small number of cross sections. In addition, cross sections of medical data are often quite complex. Shape-based interpolation is a simple and elegant solution to this problem, although it has known limitations when handling complex shapes. In this paper, the shape-based interpolation paradigm is extended to interpolate a surface through sparse, complex cross sections, providing a significant improvement over our previously published maximal disc-guided interpolation. The performance of this algorithm is demonstrated on various types of medical data (X-ray computed tomography, magnetic resonance imaging and three-dimensional ultrasound). Although the correspondence problem in general remains unsolved, it is demonstrated that correct surfaces can be estimated from a limited amount of real data, through the use of region rather than object correspondence.

A novel method is described for obtaining superior classification performance over a variable range of classification costs. By analysis of a set of existing classifiers using a receiver operating characteristic (###)curve,a set of new realisable classifiers may be obtained by a random combination of two of the existing classifiers. These classifiers lie on the convex hull that contains the original ### points for the existing classifiers. This hull is the maximum realisable ### (#####).

Volume intersection algorithms are used to reconstruct incomplete objects from their silhouettes. An imagined light source is moved about the data and the cumulative amount of "light" seen at each point an space is interpreted as indicating the likelihood that the point is inside the object. The object data need not be uniformly distributed nor exclusively surface data. Explicit distinction between noise, surface and interior data is avoided. The novel concept of a localised viewing region is introduced to overcome the inherent inability of volume intersection algorithms to reconstruct concave surfaces. Algorithms for 2D pixel and 3D voxel data are described and applied to 3D ultrasound data.

Feedback neural networks enjoy considerable popularity as a means of approximately solving combinatorial optimization problems. It is now well established how to map problems onto networks so that invalid solutions are never found. It is not as clear how the networks' solutions compare in terms of quality with those obtained using other optimization techniques; such issues are addressed in this paper. A linearized analysis of annealed network dynamics allows a prototypical network solution to be identified in a pertinent eigenvector basis. It is possible to predict the likely quality of this solution by examining optimal solutions in the same basis. Applying this methodology to traveling salesman problems, it appears that neural networks are well suited to the solution of Euclidean but not random problems; this is confirmed by extensive experiments. The failure of a network to adequately solve even 10-city problems is highly significant.

This paper presents a new algorithm for non-linear regression. It involves combining a modified regression tree with a CMAC network in a way which retains the most desirable properties of both these algorithms. The new algorithm is compared with a conventional regression tree as described in the CART book (Breiman et al., 1984). It consistently performs better on the Boston Housing Data task. The CMAC consists of a fixed non-linear mapping followed by a single layer of adaptive links. The non-linear mapping has uniform sensitivity across the whole of the input space. The essence of the new hybrid algorithm is to use a regression tree to produce a more efficient design for the CMAC non-linear mapping.

The often disappointing performance of optimizing neural networks can be partly attributed to the rather ad hoc manner in which problems are mapped onto them for solution. In this paper a rigorous mapping is described for quadratic 0-1 programming problems with linear equality and inequality constraints, this being the most general class of problem such networks can solve. The problem's constraints define a polyhedron P containing all the valid solution points, and the mapping guarantees strict confinement of the network's state vector to P. However, forcing convergence to a 0-1 point within P is shown to be generally intractable, rendering the Hopfield and similar models inapplicable to the vast majority of problems. A modification of the tabu learning technique is presented as a more coherent approach to general problem solving with neural networks. When tested on a collection of knapsack problems, the modified dynamics produced some very encouraging results.

Abstract A real‐time prototype speech recognizer has been implemented on a 66‐processor distributed‐memory parallel computer. Simple phrases are recognized in approximately 4 to 10 seconds. Scalability, performance and flexibility are the three main aims of this implementation, the ultimate goal being to construct a large vocabulary speech recognizer which responds quickly. A set of three techniques is investigated in this implementation: asynchronous methodology to minimize synchronization overheads, distributed control to avoid a central communications bottleneck, and dynamic load balancing to provide a flexible response to an unpredictable computational load. The effect on memory, processor time allocation and communications is observed in real‐time using hardware monitoring aids.

Four types of neural networks which have previously been established for speech recognition and tested on a small, seven-speaker, 100-sentence database are applied to the TIMIT database. The networks are a recurrent network phoneme recognizer, a modified Kanerva model morph recognizer, a compositional representation phoneme-to-word recognizer, and a modified Kanerva model morph-to-word recognizer. The major result is for the recurrent net, giving a phoneme recognition accuracy of 57% from the si and sx sentences. The Kanerva morph recognizer achieves 66.2% accuracy for a small subset of the sa and sx sentences. The results for the word recognizers are incomplete.>